ddflow

A portable, agent-agnostic work-queue kernel for AI coding agents.

You keep a queue of phases and tasks with declared dependencies. You say "implement phase P2". Independent tasks fan out to parallel agents in isolated git worktrees; dependent ones wait. Every task passes a quality pipeline whose gates cannot be passed by assertion. If an agent crashes, its work is found rather than lost. If everything except the log is destroyed, the project's decision history rebuilds from the log alone.

One dependency beyond python3 and git (Jinja2, for the prompt templates; see Extending it by writing text, not code). Works with Claude Code, Gemini CLI, Codex, Copilot, Cursor, Kimi, opencode, Aider, a CI job, a Makefile, or a human at a terminal — over a CLI and an MCP server that are the same implementation.

Instruction for an agent reader

If you are an AI agent and were asked to read this README, this section is your shorthand. Everything below it is detail.

What ddflow is. A work queue with rules the tools enforce, for projects that agents build. Work is phases and tasks with dependencies and declared file globs. Each task gets a lease and its own git worktree, then passes a gate pipeline: a check that could not run is unavailable, never passed; one reviewer must be from another model family; a bug closes only with a regression test that failed first. Everything is recorded in an append-only log in .ddflow/events/, committed with the code. ddflow writes none of your project's code — you do; it adds only its own config, a rules block and git hooks. Every operation is both a shell command (ddflow …) and an MCP tool (ddflow_…).

If the project already uses it (.ddflow/ exists): call ddflow_brief (or run ddflow brief) first and follow it. It replaces reading the project's rule and lesson files. Then: ddflow_next → ddflow_claim → work only in the worktree it returns → ddflow_gate_status, satisfy each gate → ddflow_merge → ddflow_complete. The full driver is docs/ddflow/drivers/implement-phase.md in that project.

To onboard a project, ask the user first: it installs a tool, registers an MCP server and writes files in their repository.

1. Install: uv tool install git+https://github.com/delian/ddflow-mcp (not on PyPI yet).
2. In the project root: ddflow adopt --agents claude (your agent's name; see Wiring it into your agent). It writes .ddflow/, the MCP registration, the commit hook and a rules block in AGENTS.md. Restart the agent session so it loads the MCP server.
3. ddflow import lists the plans, lessons, decisions and unmerged branches the repository already has, and writes nothing. Run ddflow import --apply only once the user agrees with that list.
4. ddflow config --set gate.unit_tests.command "<their test command>". Then ddflow companions lists the tools the gates expect that are missing: propose them, never install one unasked.
5. Put the user's plan in the queue (ddflow phase add …, ddflow task add … --globs …) and finish with ddflow doctor.

Exit codes, everywhere: 0 ok · 1 failure · 2 could not run / nothing to do · 3 refused. Never treat 2 as 0.

Introduction

The problem it solves

An AI coding agent is good at a task and weak at a project. One agent in one session mostly works. Run it for weeks, or run three at once, and the same failures come back:

• Work disappears. A session crashes or is closed mid-task, and the half-finished change sits in a directory nobody remembers.
• Agents collide. Two of them edit the same file, and the second merge quietly undoes the first.
• Checks that never ran look like checks that passed. The linter was missing, the reviewer endpoint was down, the tests were "run" in a summary. The agent reports done, and nothing on record says otherwise.
• The project forgets. Last week's hard-won lesson, the reason behind a design choice, the bug that was already fixed once — gone at the next session, or at the next context compaction in this one.
• Nobody can say what happened. Which instruction led to which change, and which review looked at it, lives in a chat transcript that no longer exists.

ddflow is the layer between you and your agents that makes those failures structurally hard rather than a matter of discipline. It does not write code and it is not an agent. It is a queue, a set of rules the tools enforce, and a log of everything that happened.

What changes for you

Who it is for

• One developer with one agent. A plan that survives the session, a memory that survives compaction, and a record of which checks really ran. The queue is useful even with no parallelism at all.
• Several agents in parallel — subagents, several terminals, several vendors. The dependency graph says which tasks are independent, worktrees keep them apart, and the merge step lands them without anyone switching the main checkout's branch.
• A team or a CI pipeline. The log is committed with the code, so a fresh clone knows the queue and its history. Read commands have a machine-readable --json form and every command returns the same four exit codes, so a Makefile or a CI job can drive it exactly as an agent does.

It works with the agent you already use, because everything it does is reachable both ways: as a shell command and as an MCP tool. Use whichever your agent, script or CI job has. Your workflow is text, not code — the gate pipeline, the reviewer instructions and the agent-facing prompts are files in your repository that you can edit.

What it is not

• Not an agent or a model. Your agent does the work; ddflow decides what may start, checks what was claimed, and remembers.
• Not a hosted service. Everything is files in your repository and a disposable local cache. No account, no server to run beyond the local MCP process.
• Not a replacement for your tests or CI. It runs the commands you configure and records their real exit codes and output.

A first run

# ddflow-mcp is not on PyPI yet; until the first release, install from the repository:
uv tool install git+https://github.com/delian/ddflow-mcp   # or: pipx install git+https://github.com/delian/ddflow-mcp
cd /path/to/your/project
ddflow adopt                        # registers the MCP server with your agents, writes .ddflow/ and a block in AGENTS.md
ddflow phase add P1 --title "Password reset"
ddflow task add P1.T1 --phase P1 --title "Reset-token endpoint" --globs 'src/auth/**'
ddflow next                          # what can start now, and why the rest is blocked

adopt registers the server you just installed, by its full path: an install that did not come from a package index (from git, a local directory or an archive) carries a direct_url.json in its metadata (PEP 610), and for one of those adopt writes the ddflow-mcp installed beside its interpreter rather than uvx ddflow-mcp, which would fetch from PyPI. Only an install from an index gets uvx. --launch python still forces the interpreter-plus-PYTHONPATH form.

Then tell your agent "implement phase P1". The driver adopt installed tells it to start with ddflow_brief, claim the task, work in its own worktree, satisfy each gate and land the change. ddflow cannot make an agent follow instructions, but it makes skipping them visible: the commit hook adopt installs flags a commit made without a lease (or refuses it, if you set [enforce].commit_without_lease = "block"), and ddflow complete refuses an item whose gates carry no outcome. Watch it with ddflow board, and ask ddflow doctor at any point whether the project is healthy.

How do I…?

Every row is a command you can run in a terminal and a tool an agent can call over MCP — the same implementation, so neither drifts from the other.

Every read command takes --json. Every exit code means the same thing everywhere: 0 healthy · 1 real failure · 2 could not run / nothing to do · 3 coordination refused. 2 is never collapsed into 0 — "nothing is ready" and "everything is fine" are different facts, and an agent that cannot tell them apart invents work.

Table of contents

• Instruction for an agent reader
• Introduction
• How do I…?
• Help: what it can do, and the workflow
• The default workflow at a glance
• Wiring it into your agent
• From plan mode to the queue
• When a companion is missing
• What is automated, and what is not
• Two ways to drive it
  • Standalone: a terminal, a Makefile, CI
  • As an MCP server
  • What goes in AGENTS.md / CLAUDE.md
• The workflow, and changing it
• The rules, and where each came from
• Why it is built this way
• Install into any project
  • Docker — for operators with no Python toolchain
  • Extending it by writing text, not code
  • Publishing and registry
  • Cutting a release
  • What CI checks
  • Any LLM as a reviewer — local, remote, SaaS, or a CLI
  • Companion tools
• Adopting a project that already has history
  • Verifying an import, at any time
• The model: phases, tasks, dependencies, globs
• Work that changes shape while you do it
• Architectural decisions
• Recall — "have we been here before?"
  • Similar — "is this already filed?"
• Status, progress, and loops
• The task pipeline
  • Proving a gate can fail at all
• The phase pipeline
• Human approval: a gate the agent cannot clear
• Parallelism and coordination
• Gitflow, pull requests and version tags
  • Several release lines: fixes to older majors
  • Environment branches: promoting downstream
  • Workflow choices: asked, recorded, defaulted on the record
• Many agents, one server: identity, state and sharing
  • Is it stateless?
  • Who is calling?
  • Can one server serve several projects?
  • Locking, contention and measured cost
• Crash recovery
• Reconstruction from logs alone
• Lessons that check themselves
• Checking that the checks are working
• Reading the log, and why it is never compacted
• Lessons, research and bugs
• Cadences
• Exporting documents
• Keeping session-start cost flat
• Agent portability
• Keeping the two surfaces honest
• Command reference
• Configuration
• Testing
• Documentation index

Help: what it can do, and the workflow

$ ddflow help                 # what this is, the loop, every capability grouped
$ ddflow help workflow        # workflow · import · gates · parallel · memory · recovery · config

Reachable as ddflow_help over MCP, and that is the point: an agent connecting had 59 tool descriptions and a state-aware handshake, neither of which answers "what is this, and how am I meant to work here". A tool description explains one tool to someone who already picked it; the handshake describes this repository right now.

Two halves, deliberately:

• The narrative is a template under ddflow/templates/prompts/help/, so ddflow prompts eject-style overriding applies — put your own .ddflow/prompts/help/workflow.md in place and the tool teaches your workflow.
• The capability inventory is generated from the live tool table. A hand-kept command list in a second place is the documentation-drift class, and this project has paid for it twice.

Three ratchets keep the prose honest, because a page recommending a flag that was renamed is worse than no page — whoever finds nothing reads the code, and whoever finds a wrong answer trusts it. Every command a page names must exist as a CLI leaf or an MCP tool; every topic the index offers must resolve; and every tool must fall into a group, so a new capability has to be classified rather than quietly dropped from an inventory that claims to be complete.

The default workflow at a glance

Three pictures of what ships by default. Every step is configurable (see The workflow, and changing it); ddflow workflow prints what this project actually runs.

The agent's loop. One item at a time: pick it, lease it, clear its gates, land it. Each arrow labelled with an exit code is what the tool says, not a convention — 2 means nothing to do, 3 means refused, and neither is ever read as success.

flowchart TD
    S["Session start<br/>ddflow brief · recover"] --> R{"Crashed agent's<br/>work left over?"}
    R -- yes --> SV["Inspect the worktree, salvage,<br/>ddflow release"] --> N
    R -- no --> N["ddflow next"]
    N -- "exit 2: nothing ready" --> W["ddflow wait<br/>sleeps until a holder lets go"] --> N
    N -- "exit 0: items ready" --> C["ddflow claim<br/>lease + isolated git worktree"]
    C -- "exit 3: refused<br/>(lease or file-glob conflict)" --> N
    C --> P["Task pipeline<br/>satisfy every gate in order"]
    P --> M["ddflow merge<br/>lands the branch from the primary checkout"]
    M --> D["ddflow complete<br/>checks every gate has an outcome<br/>+ a different-family review"]
    D -- "exit 3: refused" --> P
    D -- done --> N

With [flow].integration = "pr", merge opens a pull request and parks the item in review instead; next completes it when the request merges.

The task pipeline. Ten gates, in order. Heavy borders are the gates gates.required names by default (implement, unit_tests, merge); the rest still need some outcome — passed, failed, unavailable, partial, or an explicit skip with a reason — because silence is not a pass.

flowchart LR
    subgraph A["You, the agent"]
        direction TB
        g1["1 · research<br/>falsifiable claim + probe"] --> g2["2 · rules<br/>ddflow brief --item"] --> g3["3 · implement<br/>in the item's worktree"]
    end
    subgraph X["A different model family"]
        direction TB
        g4["4 · rubber_duck<br/>try to refute it"] --> g5["5 · critic<br/>diff vs. stated intent"]
    end
    subgraph T["Tooling"]
        direction TB
        g6["6 · standards<br/>linters, architecture"] --> g7["7 · unit_tests<br/>the suite, executed"]
    end
    subgraph B["You, again"]
        direction TB
        g8["8 · bug_hunt<br/>recurring bug classes"] --> g9["9 · dedupe<br/>already exists?"]
    end
    g10(["10 · merge<br/>ddflow lands it"])
    A --> X --> T --> B --> g10

    classDef req stroke-width:3px
    class g3,g7,g10 req

The phase pipeline. A phase wraps its tasks and checks the whole before it lands:

flowchart LR
    p1["research<br/>the phase's open questions"] --> p2["tasks<br/>each runs its own<br/>ten-gate pipeline,<br/>in parallel where<br/>globs allow"]
    p2 --> p3["unit_tests"] --> p4["bug_hunt"] --> p5["dedupe"]
    p5 --> p6["live_test<br/>run the real thing"] --> p7["corrections"] --> p8["docs<br/>README and docs<br/>match the change"] --> p9(["merge"])

Details: The task pipeline · The phase pipeline · Parallelism and coordination · Crash recovery.

Two ways to drive it

The CLI is the whole product. The MCP server is a second surface over the same commands, and tests/test_mcp_parity.py fails if the two diverge — every subcommand has a tool, every flag is reachable, and each exemption carries a written reason.

Standalone: a terminal, a Makefile, CI

$ ddflow init
$ ddflow config --set gate.unit_tests.command "python -m pytest -q"
$ ddflow phase add P1 --title "Billing" --globs "src/billing/**"
$ ddflow task add P1.T1 --phase P1 --title "Tax rules" --globs "src/billing/tax.py"

$ ddflow next                          # exit 2 = nothing actionable; exit 1 = unknown --phase
$ ddflow claim P1.T1                   # exit 3 = refused, with the reason
leased P1.T1 · worktree .ddflow-worktrees/P1.T1 · branch ddflow/P1.T1

$ cd .ddflow-worktrees/P1.T1 && ...    # do the work
$ ddflow gate status P1.T1             # what the pipeline wants next
$ ddflow gate run P1.T1 unit_tests     # runs it; the exit code IS the evidence
$ ddflow gate record P1.T1 implement --outcome passed --evidence "added tax.py"
                                       # failed/unavailable/partial/skipped also need --reason
$ ddflow complete P1.T1                # exit 3 lists whatever is unsatisfied
$ ddflow merge P1.T1

You get everything except the judgement. Command gates run themselves; agent gates wait for a human to record an outcome, and ddflow gate skip <id> <gate> --reason "..." is the escape hatch — recorded as a skip, never as a pass.

Changelog line (optional). ddflow complete P1.T1 --changelog "Added: tax rounding" (and ddflow bug fixed B1 --regression-test ... --changelog "Security: ...", MCP changelog on ddflow_complete / ddflow_bug_fixed) records one line for the project's changelog on the completion event. The category is one of Added, Changed, Deprecated, Removed, Fixed, Security (case-insensitive; anything else is refused with that list); --changelog skip (or internal) marks work that must not appear in the changelog. complete never asks for it and never requires it, and logs without it fold exactly as before.

In CI, the exit codes are the interface:

check:
	ddflow doctor        # 1 = integrity problems, each named
	ddflow workflow      # 1 = the pipeline does not hang together
	ddflow cadence       # 2 = no periodic pass is due

2 is never "no problem". A job that treats it as success reports a green build for a suite that never ran.

As an MCP server

ddflow mcp speaks newline-delimited JSON-RPC over stdio, and treats the directory it was started in as the caller's location (as ddflow-mcp does), so a merge from inside a worktree keeps that worktree. You rarely run it by hand — ddflow adopt writes the launch entry into each agent's own config and leaves existing servers alone:

22 agents are supported. The full table, with what each one gets, is in Wiring it into your agent.

It also copies the driver to docs/ddflow/drivers/, and installs the pre-commit hook that enforces claim-before-you-edit.

What an agent sees the moment it connects, with no call to make:

• Instructions, returned inside the initialize result itself — and state-aware: what is ready, what is in flight, which setup is missing, whether this project has history worth importing, whether an import was left unfinished.
• Tools — one per CLI command.
• Resources — ddflow://board, ddflow://brief, ddflow://lessons, ddflow://research.
• Prompts — which a client turns into slash commands. Tools are things an agent calls; prompts are things you invoke.

A refused call says so first. Over MCP a tool that did not do what was asked — a claim refused for an overlap (exit 3), or any call whose result would otherwise be its success shape in nulls — returns a JSON body whose first key is "refusal": {"reason": ..., "outcome": ..., "exit": ...}, followed by whatever the operation actually said (a refused claim's alternatives). The lead is named refusal for any call that did not do what was asked: outcome (failed, nothing or refused) and exit say which, so an exit-1 failure is not mistaken for a refusal. A tool whose body is one field (ddflow_decision_show on an unknown id) leads with the same object instead of a bare null. Exit 2 ("nothing") keeps its declared keys after that lead, and any other data it carried; a data field that is itself named refusal is kept as refusal_data. A result that fills its declared shape is left as the CLI's --json prints it, with the reason in the second content block.

Two tools exist so an agent can orient itself without being told: ddflow_help (what is this, what is the loop) and ddflow_workflow (what are the rules here).

What goes in AGENTS.md / CLAUDE.md

ddflow adopt writes it as a managed block between <!-- DDFLOW:BEGIN --> and <!-- DDFLOW:END -->. Your own prose around it is preserved; re-running updates only what is inside. If you write it by hand, four things have to be in it:

1. Start every session with ddflow_brief (or ddflow brief in a shell).
2. Claim before you edit — ddflow_next → ddflow_claim → work in the worktree it creates.
3. The loop — ddflow_gate_status → satisfy each gate → ddflow_complete → ddflow_merge.
4. The exit codes, and that 2 is not success.

Without that block an agent sees the tools and has no reason to reach for them before editing. The block is what makes the queue authoritative rather than optional — and it is 232 words, because an instruction file nobody finishes reading is one nobody follows.

The workflow, and changing it

$ ddflow workflow
# The workflow this project runs

   1. research      agent
   2. rules         agent
   3. implement     agent       (required)
   4. lint          command     (required, NOT proven able to fail)
      $ ruff check .
   ...

## The rules, and where each came from

  gates.require_outcome                  True              [default]
  gates.enforce_order                    block             [file]
  schedule.max_parallel_tasks            4                 [default]

One answer to "what are the rules here": every gate in order, which are commands and which you perform, which are required, which need evidence, which need a different-family reviewer, which have been proven able to fail — plus the completion rules, the caps, the reviewers, and where each value came from, so a deliberate choice is distinguishable from a default nobody touched.

Changing it

$ ddflow workflow gate lint --command "ruff check ." --into task --after implement --required
$ ddflow workflow pipeline task research,implement,lint,unit_tests,merge
$ ddflow workflow drop dedupe

All four reach MCP — ddflow_workflow, ddflow_workflow_pipeline, ddflow_workflow_gate, ddflow_workflow_drop — so an agent can change the workflow with the operator's agreement. Their descriptions say to ask first and offer dry_run, because a pipeline governs every future item, not the one in hand.

Nothing is written until it is checked, and the order is the point: compose the change, validate the result, then replace the file atomically.

• A pipeline naming an undefined gate is refused, naming the near miss. That one is otherwise silent and permanent: the outcome folds to empty, completion refuses it forever, and gate record rejects the id as unknown — so the item can never be completed at all, and nothing says why.
• An unknown section or knob is refused, with a suggestion. [gatez] is valid TOML and used to be written happily, breaking every later command — the write path validated the merged text for syntax and then validated the config already on disk, which is a writer checking the state it is replacing.
• Dropping a gate takes it out of required too, or it becomes a requirement that quietly requires nothing.

ddflow workflow and ddflow doctor both re-run those checks against what is on disk. Everything is a file you can also edit by hand: gates in [gate.<id>], reviewers in [[reviewer]], companions in .ddflow/companions.toml, and every prompt — including the instructions your agent receives at connect — under .ddflow/prompts/.

One caveat with MCP: the connection instructions are computed once, when the server starts. A workflow changed mid-session is live for every tool call immediately, but the text the agent was handed is stale. Tell it to call ddflow_workflow, or restart.

Why it is built this way

The append-only event log is the source of truth; everything else is a projection that can be deleted and re-derived. The SQLite index, the markdown boards, the search index, the recovery bundle — all disposable, all rebuilt by ddflow rebuild.

That single inversion is what makes the four hard properties fall out for free rather than needing to be engineered:

The design decisions, with the probes that settled each, are in docs/RESEARCH.md. The two that most shaped it:

• SQLite-on-NFS is correct here but 36× slower than local (measured, 12 processes × 40 increments). So the log is authoritative and the database is a disposable cache — which also happens to be the choice that stays correct on filesystems where locking is broken.
• An expired lease must never be reclaimed automatically. A crashed agent's worktree is sometimes irreplaceable work and sometimes a superseded draft, and nothing in the metadata distinguishes them. Recovery measures and advises; it never deletes.

Install into any project

One line in your agent's MCP config. Nothing else.

Until the first release is on PyPI, uvx ddflow-mcp has nothing to fetch: install from the repository and run ddflow adopt, which registers that installation instead, as in A first run.

{ "mcpServers": { "ddflow": { "command": "uvx", "args": ["ddflow-mcp"] } } }

uvx fetches and runs the published package in an ephemeral environment on first use — no clone, no virtualenv, no PYTHONPATH, no install step for an operator to forget, and no vendored copy to drift from upstream. ddflow needs one runtime dependency beyond python3 and git (Jinja2), which is what lets it install inside sandboxes, CI images and other tools' ephemeral containers.

Then, from the agent, with no shell at all:

That is the whole adoption. The per-project instruction text is 232 words — a managed block in AGENTS.md, because the MCP tool descriptions already carry the how, and a second copy of that would drift from the one the model actually reads.

uv tool install ddflow-mcp        # or: pipx install ddflow-mcp
cd /path/to/your/project
ddflow adopt            # every supported agent
ddflow adopt --agents claude,cursor,vscode,kimi   # or name the ones you use

Docker — for operators with no Python toolchain

{ "mcpServers": { "ddflow": { "command": "docker", "args": [
    "run", "-i", "--rm",
    "-v", "${workspaceFolder}:/repo",
    "--add-host=host.docker.internal:host-gateway",
    "ghcr.io/delian/ddflow-mcp:latest" ] } } }

ddflow adopt --launch docker writes exactly that. The image is 107 MB (Alpine; ddflow is pure standard library, so there is no compiled dependency to worry musl about) and behaves identically on Linux, macOS and Windows.

Four things go wrong when a containerised tool touches a bind-mounted git repo. All four are silent, one of them loses work, and all four are handled:

And one that cannot be fully handled, so it is reported: 127.0.0.1 inside a container is the container. A model server on your own machine is not reachable from there. ddflow rewrites loopback reviewer URLs to host.docker.internal, and ddflow doctor tells you that on Linux you must also pass --add-host=host.docker.internal:host-gateway, because unlike Docker Desktop the Linux engine does not provide that name.

The related portability fix: worktree paths are stored in the event log relative to the repo root. The log is committed and shared, so an absolute path is true only on the machine that wrote it — false for a teammate who cloned elsewhere, for CI, and for a container where the repo is /repo. Pinned by test_the_event_log_carries_no_absolute_paths.

Extending it by writing text, not code

Every prompt is an external template, resolved config → project → shipped:

ddflow prompts list              # where each template currently comes from
ddflow prompts eject             # copy the shipped ones into .ddflow/prompts/
$EDITOR .ddflow/prompts/review_system.md

Adding a mode of your own: [[macro]]. Overriding a shipped workflow needs no code, and neither does adding one. A macro is a named, parameterised prompt — "enter debugger mode" — that appears everywhere the shipped workflows do: prompts/list and prompts/get over MCP, which is what a client turns into a slash command, and ddflow prompts list|show in a terminal.

# .ddflow/config.toml   (or .ddflow/macros.toml, if you prefer to split it out)
[[macro]]
name = "debugger"
title = "Enter debugger mode"
description = "Reproduce first, then bisect. No fix without a failing probe."
params = ["symptom"]                                  # required, not optional
tools  = ["ddflow_bug_found", "ddflow_gate_run", "ddflow_bug_fixed"]
prompt = """
You are debugging: {{ symptom }}

Reproduce it before you theorise. Paste the command and its output.
"""

Use prompt_file = "docs/modes/debugger.md" instead for anything long enough that TOML quoting gets in the way.

When to reach for a macro rather than a gate. A gate is a step every item passes through, recorded against that item and blocking its completion. A macro is a MODE an operator enters, belonging to no item and recorded nowhere — "audit this release", "handle this incident". If the thing should hold up a task until it is done, it is a gate; if it is a way of working you want to name and re-enter, it is a macro. Putting a mode in the pipeline makes every task wait for something that was never about that task.

tools is declarative, not a sandbox. It is rendered into the prompt as the ordered set the mode expects, so the agent is told what the mode is for and the next reader can tell what it was supposed to do. It does not restrict what the agent may call — MCP has no mechanism for that, and claiming a security property this cannot honour would be worse than not having it. This is the deliberate departure from dx-zero/mcpn, whose toolMode: situational lets the model pick freely from a bound set with no recorded ordering: a session you cannot replay is a session you cannot review, which is the property the event log exists to give you.

Three things a macro refuses, because each alternative fails quietly: a missing parameter (a prompt with a hole in it reads as a complete instruction), a name that belongs to a shipped command (silent shadowing leaves you editing a block that does nothing), and both prompt and prompt_file (two sources for one body means one is dead and looks live). A refused macro is refused alone and by name — the others still load — and ddflow prompts list, prompts show, doctor and the MCP prompts/list say which one and why; an undecodable prompt_file is a named problem too. When the whole [[macro]] config cannot be read (a parse error, an unknown field), MCP prompts/get for an unknown name says why (not loaded: ...) instead of only "unknown prompt".

Including the one the agent actually reads first. mcp_instructions.md is the block an MCP client injects into the model's context on connect — the workflow, the reporting duties, and which companion tools to reach for. It is the file to edit when you want this project to work differently:

ddflow prompts eject mcp_instructions
$EDITOR .ddflow/prompts/mcp_instructions.md      # or [prompts] mcp_instructions = "..."

It renders against the live state — adopted, task_pipeline, setup_todo, companions, missing_companions, gate_gaps, recoverable, loops — so the instruction is the next concrete action rather than a fixed blurb the model learns to skip. A broken override says so in the instruction block itself instead of falling back to the default: this is the one surface where nobody would ever notice their edit was not live.

Templates render with Jinja2, which is ddflow's one runtime dependency, and with a strict standard-library renderer when it is absent — a stripped deployment with no reachable package index still starts. The shipped templates use the subset both engines agree on, and tests/test_template_engines.py walks the template REGISTRY, rendering every entry through both engines and asserting the outputs are byte-identical.

That test is iterated rather than hand-listed for a reason. Its predecessor named three templates in a dict, mcp_instructions.md was never added, and in 0.1.1 the largest and most important template rendered correctly under Jinja2 and failed under the fallback — so the entire MCP handshake for an unadopted repository, the first thing a new user ever sees, degraded to ddflow's instruction template could not be loaded. Jinja2 was not a declared dependency at the time, so developers had it and the project venv did not: python -m pytest was green and uv run pytest was red on the same commit.

The fallback now raises on any construct it does not implement rather than copying it through. The old regex engine emitted what it could not parse, so a condition as ordinary as {% if a or b %} — which its single-name pattern never matched — reached the client as literal template source.

Both renderers are strict about undefined variables: a prompt silently missing the diff it was supposed to carry is the vacuous review in template form — the model dutifully reviews nothing and reports no findings.

The rest is TOML: gates and their pipelines ([gate.*], gates.task_pipeline), reviewers ([[reviewer]]), companions ([[companion]]), enforcement ([enforce]), cadences, and the rest of the 150 knobs. ddflow config --set <key> <value> edits one key in place, preserving comments.

What is committed, and what stays on your machine

ddflow recommends services; it never ships one person's configuration. Two layers:

The local files are read last, so they win; ddflow config --explain reports such a value's source as local. Every writer of an operator-specific value targets the local layer by default:

ddflow reviewers add --preset ollama --model qwen3:8b   # -> .ddflow/local/reviewers.toml
ddflow reviewers detect --write                         # -> .ddflow/local/reviewers.toml
ddflow config --local --set gate.unit_tests.command "pytest -q -n 48"
ddflow config --local --append-toml "$(cat my-reviewer.toml)"

--shared on reviewers add / reviewers detect --write commits the block to .ddflow/config.toml instead — only for a service every clone reaches at the same address. config --set and --append-toml stay committed unless you pass --local (over MCP: ddflow_configure with local=true, ddflow_reviewers_detect with shared=true), because a test command or a pipeline is project policy. The human-gate guards hold on both layers: no writer sets gate.<id>.human, and none can drop a human gate from a pipeline. .ddflow/local/ carries its own * .gitignore, so it stays uncommitted even in a project whose .ddflow/.gitignore predates it. API keys are never written anywhere — only the name of the variable that holds one.

A fresh ddflow init says the same in the config.toml it writes: the commented [[reviewer]] example points at .ddflow/local/reviewers.toml (not at config.toml), and a short comment block separates what is committed (generic project policy) from what is local (your endpoints, hosts, key variable names, machine sizing: ddflow config --local).

Publishing and registry

Nobody should have to paste JSON into an IDE to use this. server.json is the MCP registry manifest (io.github.delian/ddflow-mcp), and publishing it is what makes ddflow findable in the VS Code and Cursor marketplaces rather than something you configure by hand. It offers three ways to run the same server, so a client picks whichever it supports:

The image is built for amd64 and arm64, because an Apple-silicon operator running it under emulation pays that cost on every tool call, and tool calls are all this server does.

How CI authenticates — four mechanisms, one stored secret:

Docker Hub is the only one that needs a long-lived credential, because it has no OIDC equivalent. Use an access token scoped to this repository, never an account password. If that is one secret too many, delete the Docker Hub login and its two tags — ghcr.io alone satisfies the OCI entries a marketplace needs, and server.json lists both so a client picks whichever resolves.

environment: release on the publishing jobs is a control worth knowing about: point it at a GitHub environment with required reviewers and every release waits for a human, with no change to the workflow.

Order matters and the workflow encodes it. mcp-publisher validates that every package named in the manifest exists, so the registry step runs after both PyPI and Docker — publishing the manifest first would advertise a version nobody can fetch.

The registry verifies ownership, and the workflow checks it first. The registry accepts a package only when the artefact itself names the server: the PyPI package's README carries <!-- mcp-name: io.github.delian/ddflow-mcp --> (the first lines of this file), and each image carries the label io.modelcontextprotocol.server.name with the same name. The verify job runs tests/test_registry_ownership.py first, so a release the registry would reject — a missing marker, or a label that does not match — is refused before anything is uploaded; a server.json description over the registry's 100-character limit fails the suite the same way. So do the registry's per-package rules, which the JSON schema does not express and the registry enforces only at publish: an oci package must not carry a version field (or registryBaseUrl/fileSha256) — the tag in its identifier is the version — while the pypi package must carry one. publish #40 failed on exactly that, at the last job, after PyPI and both images had shipped; tests/test_registry_manifest_rules.py now encodes the rules offline and verify runs it first. The publish step also retries a transient registry failure (a 504, a 408/429, a network error) with backoff, checking the registry for the exact version after every attempt, because the publish behind a 504 may have committed; a 4xx that retrying cannot fix fails at once, and only after the budget does it fail with a "Re-run failed jobs" hint.

Four things gate a release, and each exists because the failure it catches is public and irreversible:

• the tag, ddflow/__init__.py (the one declared version), server.json's version and every OCI identifier's tag must agree — a :0.1.0 left behind while version moved on publishes a manifest pointing at the previous image, installable and wrong;
• the full suite, plus the slow end-to-end scenarios, which -m 'not slow' otherwise excludes from every ordinary run;
• the wheel must install into a clean venv and run, and carry its templates — uv build succeeding proves the metadata parses, not that ddflow help works;
• the image must answer initialize over stdio. A built image that cannot is a broken release every marketplace will happily offer.

Cutting a release

$ git push origin main           # that is the whole release

Every push to main that changes shipped code releases, with the PATCH version bumped. Major and minor move only when you move them. "Shipped" means ddflow/, pyproject.toml, uv.lock, Dockerfile, docker-entrypoint.sh, .dockerignore, server.json, server.template.json or README.md (the PyPI page, and the line the registry verifies): a push of other docs, tests or the ddflow event log releases nothing, because PyPI keeps every version forever and one identical to the last is noise nobody can withdraw. CI runs scripts/bump.sh patch, commits release 0.1.2 to main, publishes PyPI, Docker Hub, ghcr.io and the MCP registry, then creates v0.1.2 and a GitHub release — last, and only once every publish succeeded, because a tag pointing at a half-release is worse than no tag: it looks authoritative.

Pull after a release. The release commit is CI's, so your main is one commit behind it and the next push is refused until you git pull.

How the gate picks the version: it publishes the declared version if PyPI does not have it yet, and bumps patch only if it does. So moving major or minor is yours to do —

$ scripts/bump.sh minor          # 0.1.4 -> 0.2.0 (or: major, or an exact 1.0.0)
$ git commit -am 'release 0.2.0' && git push origin main

— and CI publishes exactly 0.2.0; the next push that bumps nothing releases 0.2.1.

The same rule makes a release that failed before its PyPI upload retry its number with the next push. One that failed after it — Docker Hub, ghcr.io, the MCP registry — does not: PyPI has the number, so the next push moves past it; re-run that run's failed jobs from the Actions page instead. A bump never lands on a number PyPI already holds (a v* tag can publish one out of band): CI skips to the next free patch before it commits anything. The bump is pushed to main before anything publishes: a push that loses a race with another commit fails the run and publishes nothing, where pushed last it would leave PyPI holding a version main does not declare. Runs are serialized, and only main or a v* tag releases.

The version is declared in one place: __version__ in ddflow/__init__.py, the only line a bump edits. pyproject.toml reads it (a hatch dynamic version, so uv.lock records no version for the project and a bump never makes the lock stale), SERVER_INFO (what the server tells every client it is, and what ddflow --version prints) is built from it, and server.json — the manifest's version, the pypi package's version and the tag inside every OCI identifier (an oci package has no version field) — is a generated, committed file: edit server.template.json, then run scripts/render_server_json.py (--check fails when server.json is not the render; a test and the publish workflow both run it). scripts/bump.sh edits the literal, renders server.json and re-reads the result; tests/test_packaging.py fails if anything drifts. Do not edit server.json by hand.

scripts/release.sh runs all of that locally and publishes nothing. It is dry by default, needs no credentials, and exists because a tag is not reversible: PyPI refuses a re-upload, :latest is on someone's disk before you notice, and a registry manifest is what an IDE offers people. If it fails on your laptop, the tag was going to fail an hour later in public. --publish is the escape hatch for when CI is unavailable, and it makes you type the version to confirm.

What CI checks

.github/workflows/ci.yml runs on every push and pull request, in four jobs that fail for different reasons so you can tell at a glance which:

Two of those exist because of specific failures. The wheel check is there because uv build succeeding proves the metadata parses, not that ddflow help works — a wheel missing its templates fails on the user's machine. And gitleaks is there because this project has already committed a live API key: a secret in git is a leaked secret, rotation is the only remedy, so the check that matters is the one that runs before every push.

bandit deliberately skips tests/, which use subprocess and temporary paths constantly and by design. A scanner that cries wolf on every fixture is a scanner nobody reads.

The same checks run before you push. Install the pre-push hook once per clone:

$ ln -sf ../../scripts/ci/pre-push "$(git rev-parse --git-common-dir)/hooks/pre-push"

It runs what CI runs, from .pre-commit-config.yaml: ruff, the wheel probe, bandit, the dependency allowlist, gitleaks over the commits being pushed, the unit suite and the scenarios, and a schema check of the workflow files. The build probe and the allowlist are scripts/ci/ files that CI calls too, so the two cannot drift.

It runs them in a scratch worktree of the commit being pushed, not in your checkout — which is why it is not pre-commit install. pre-commit fails a hook when anything in the tree changes while it runs, and ddflow agents append to the committed event logs constantly: in your own tree the test hook failed with every test green. A scratch tree has no other writers, and it checks what you push rather than what is uncommitted.

The unit suite and the scenarios run as one pytest session, one worker per CPU (-n auto) with work-stealing, so the scenarios overlap the unit tests instead of following them: about a minute on a 192-thread machine. The scenarios take CI's no-model-server path (DDFLOW_DEMO_NO_REVIEWER=1), since a real review there costs minutes and is not what CI checks.

It is a push hook, not a commit hook, because ddflow's own hooks own the commit (ddflow hooks install), and because minutes per commit teaches --no-verify. SKIP=tests git push skips one check by id. It does not run the 3.11/3.13 matrix, CodeQL, pip-audit or the load suite.

Any LLM as a reviewer — local, remote, SaaS, or a CLI

The critic and rubber_duck gates are run by ddflow, not claimed by the agent. Point them at whatever you have:

ddflow reviewers presets            # 19 ready-made provider settings
ddflow reviewers add --preset ollama --model qwen3:8b
ddflow reviewers detect --write     # probe local ports, register what is serving (machine-local)
ddflow reviewers test               # send a known-buggy diff, check the reply

Four backends, because "any LLM" means four wire formats in practice:

command is the escape hatch that makes the answer to "can it use X?" always yes: a model with no HTTP API, behind a corporate gateway, or wrapped in an in-house tool is still usable, with no SDK and no dependency.

Who may add a reviewer. A reviewer decides whether a review counts as independent, so an agent must not be able to mint one (decision D-reviewer-trust):

• A kind = "command" reviewer runs any program and can print any verdict, so only a person adds one: by editing .ddflow/local/reviewers.toml, or with ddflow reviewers add from their own terminal. Every agent surface refuses it — ddflow_configure, config --append-toml, and any command run under --agent or DDFLOW_AGENT, or in a Claude Code shell (CLAUDECODE, the one harness marker known for certain; another harness is recognised by the --agent/DDFLOW_AGENT its ddflow setup passes).
• A reviewer a tool writes (reviewers add, reviewers detect --write, ddflow_configure) is recorded with who wrote it, and its reviews do not count toward the cross-family rule until a person checks the entry and runs ddflow reviewers approve <name> (ddflow reviewers approve alone lists what is waiting). Approval is for the entry as it was: a tool changing its endpoint, model, family or command makes a new, unapproved reviewer; tuning max_tokens or hedge does not. reviewers approve has no MCP tool and refuses under an agent identity.
• A reviewer no tool wrote — every entry configured by hand, including all of them from before this rule — counts exactly as before.

Like human gates, this makes a forged reviewer visible in the log; it cannot stop a shell edit of the local files.

[[reviewer]]
name   = "local-qwen"
kind   = "openai"
base_url = "http://127.0.0.1:11434/v1"
model  = "qwen3:8b"
family = "alibaba"              # must differ from the author's family
gates  = ["critic"]
# Optional: start it if it is not already running.
launch = { command = "ollama serve", ready_url = "http://127.0.0.1:11434/v1/models" }

[[reviewer]]
name    = "claude-via-cli"
kind    = "command"
command = "claude -p --model {model}"
model   = "claude-sonnet-5"
family  = "anthropic"
gates   = ["rubber_duck"]

Auto-launch is opt-in per reviewer — starting a multi-gigabyte model server as a side effect of asking for a code review is a surprise nobody wants by default. When it fails it never leaves a half-started process behind, because a reviewer stuck "starting" forever is indistinguishable from one that is down except that it also holds a process.

Keys are never written to the config. Only api_key_env, the name of an environment variable — the config file is committed, and a key in git is a leaked key.

Every way of not reviewing is reported distinctly, with its remedy: no key names the variable, a missing CLI names the binary, a dead server names the launch block you could add, a non-zero exit shows stderr, and empty output on exit 0 is UNAVAILABLE rather than "no findings" — the vacuous pass arriving by the most innocent-looking path there is.

Reasoning models need a large max_tokens. Default 32000, measured not guessed: on Qwen3.8-Flash-Next over a 30 KB diff, a 6000-token budget produced zero characters of content — the whole budget went to reasoning and the reply was truncated. That case is reported as TRUNCATED with the remedy named, never as an empty completion and never as a clean review. A chunk lost that way on every copy is tried once more, split in halves when it splits, before it is reported unreviewed.

Leave temperature unset unless you mean to override the model. Unset, the request carries none and the server applies the model's own recommended sampling. A reasoning model sampled cold (ddflow used to send 0.3) can loop — "let me reconsider" — until the whole budget is gone; at its vendor's recommended 1.0 the same diff reviewed fully.

Reviews run in parallel, and each chunk is raced. A large diff is reviewed in chunks (max_chunk_chars). They go out together — by default every chunk and every copy in one wave (up to 32 requests); max_concurrency caps it — so a review takes as long as its slowest chunk rather than the sum. Each chunk is also sent hedge times (default 2): the first copy that answers on contract is the chunk's review, and the rest are cancelled — the connection is closed so the server aborts the request, a command reviewer's process group is killed. A chunk whose every copy failed is reported exactly as before; nothing is counted as reviewed that was not.

Why: a reasoning model's time is its reasoning length, and that is random. Measured on a LAN vLLM: the same 3 KB diff took 64–391 s across eight identical calls, about one call in four on a hard diff never answers before max_tokens, and the server was idle — eight concurrent requests each ran 28% slower for 5.7× the throughput. A 24 KB review went from 1 274 s (chunks in sequence) to the time of its slowest chunk.

[[reviewer]]
# ...
hedge = 3             # copies per chunk; 1 turns racing off
max_concurrency = 10  # cap on requests in flight; 0 (default) = all in one wave

On a metered API a cancelled copy is still billed for what it generated before it was stopped: hedge = 1 if cost matters more than time there. On a single-slot local server (an ollama with OLLAMA_NUM_PARALLEL=1) the extra copies queue behind the first ones, which are always sent first; there, set max_concurrency to the server's slots so a queued request's timeout_s does not run out while it waits.

What a review reports, and how to finish one that was partial. Every chunk that was not reviewed is named, with its files and its cause: a chunk whose generation ran out the clock is labelled did not converge (the model was reachable), distinct from an endpoint that was down. No chunk is header-only and none is dropped; hunk-header context is stripped so a finding cannot cite a function the diff does not touch. Progress is printed per chunk, the caller's lease is renewed while the review runs, and extra_rules on a [[reviewer]] reaches its prompt. Findings are numbered #1..#N.

A review that outlives its caller keeps its findings. Each chunk's whole reply is appended to .ddflow/local/reviews/<item>.<gate>.<reviewer>.<run>.jsonl (git-ignored; one file per reviewer and run, so a re-review never overwrites an earlier one) the moment it arrives, and the gate evidence records output_file (repo-relative) and output_digest beside the per-finding text, so a call that is cut off (a long critic over MCP) loses nothing. Over MCP, ddflow_review sends notifications/progress for each progress line when the client passed a progressToken, which resets a client's idle timer.

ddflow review T1 --gate critic --chunk 2,5      # re-review only chunks 2 and 5 of that review
ddflow review triage T1 --gate critic --finding 3 --refuted --probe "tests/test_x.py::t shows..."
ddflow review triage T1 --gate critic --finding 1 --confirmed --probe "fixed in 4f2a, test_y"

--chunk re-runs the named chunks of the same cut — the same diff, max_chunk_chars and reviewer, checked before anything is sent — and merges their coverage into the gate's recorded outcome. review triage records what became of each finding as its own event (review.triaged): refuted, with the run that shows it false, or confirmed, with the fix or test that answers it. The gate's outcome is not changed — a review that reported findings stays failed, and that does not block completion; the log now shows what became of each finding. Triage appears in gate status and show, and a re-review carries it over only for a finding whose text is identical (decision D-review-triage). Over MCP it is ddflow_review_triage; --chunk is an argument of ddflow_review. The triage verb is required: --finding/--refuted/--confirmed/--probe on a plain review are refused (exit 1) before any reviewer is contacted, as is --chunk on review triage. Finding numbers are per gate, so --gate is never defaulted on review triage: omitted, it is refused (exit 1) naming the gates that have findings when several do, and resolved to that gate (the output names it) when exactly one does. ddflow_review_triage's gate works the same.

The review-round budget ([review]). A shipped default in every project, with no configuration: a gate (rubber_duck, critic) gets 2 full review rounds per item, then a third is refused (exit 3) with the way forward — later rounds each find fewer defects than the one before (measured: rounds 3 and later yielded about a quarter of the confirmed findings). What stays allowed after the cap, always:

ddflow review T1 --gate critic --delta      # recheck ONLY what changed since the head the last review covered
ddflow review triage T1 --gate critic --finding 2 --refuted --probe "..."   # settle what is left
ddflow review T1 --gate critic --force --reason "..."   # one more FULL round; the reason is recorded

A full round is any review that can cover the item's whole diff. What is not one is judged by what it covers, not by the flag: --delta, a --commit <sha> or --base <ref> at or after the head the gate's last review covered (so it can only be narrower), and a --chunk re-run of a recorded review. Those are never refused; a --base or --commit that reaches back past that head counts as a full round. Rounds are counted from the log's recorded reviews (review_kind, round, rounds and reviewed_head in the gate evidence), so a re-claim, a delta or a manual gate skip does not reset the count; a round that reached no reviewer is not counted. --delta reads the reviewed head from the log too, so it works after a re-claim or a gate skip; it is refused only when no review was ever recorded or nothing changed.

Delta re-reviews are the default. Once a gate has a recorded review that reviewed the whole diff, a plain ddflow review T1 --gate critic is a delta: it reviews only the commits since the head that review covered (reviewed_head), says so (delta review of 1 commit since a1b2c3d4e5), is not a full round and is never refused by the cap. Its findings and coverage are merged into the gate's record: earlier findings stay where they were (their #N and their triage, which is keyed by the finding's exact text, are unchanged), a byte-identical finding is not duplicated, the delta's new ones are appended and the output maps each to its place in the record; the gate's outcome is the delta's own, except that a clean delta does not pass while an earlier finding on the record has no triage verdict (it records failed and says so; review triage settles each, then the next clean delta passes). gate status shows the rounds separately (1 full round, 2 delta rounds). --full forces a full round (counted against review.max_rounds; combined with --delta it is an error). A delta falls back to a full round, and says why, when the reviewed head is no ancestor of the current head (the branch was rebased or amended) or, for the automatic delta, the earlier review was partial; a review that never reached a reviewer records no reviewed head at all, so the next delta starts from the last real one; with nothing changed since the reviewed head the review is refused and names --full. A --force --reason review is a full round, like --full (and, like --full, an error with --delta); a --chunk, --commit or --base review is not second-guessed. review.delta_default = false is the behaviour before this knob: every review a full round. Knobs, changeable at every layer:

ddflow config review.max_rounds 3                 # this project (committed .ddflow/config.toml)
ddflow config review.max_rounds 0 --local         # this machine only (.ddflow/local/config.toml)
ddflow config review.delta_default false          # every review a full round again (add --local for this machine)
ddflow review T1 --gate critic --full             # one full round, whatever the default
                                                  # (`--set KEY VALUE` is the same)
DDFLOW_REVIEW_MAX_ROUNDS=0 ddflow review ...      # one run

Over MCP, ddflow_configure accepts review.max_rounds, review.on_exceed and review.delta_default on either layer (local=true), and tells the operator in its reply and in a session note — an agent does not lift the cap for itself; --force --reason is CLI-only and ddflow_review takes delta=true and full=true. ddflow config --explain --filter review. documents the knobs. A tool built before this knob skips an unknown [review] key with a note rather than failing.

Companion tools

ddflow imposes the order and demands the evidence. It does not perform the judgement inside most of its gates: standards wants an automated standards review, research wants documentation to check a claim against. (rules wants memory of the last time somebody hit this, and ddflow serves that one itself: ddflow brief shows the lessons and the operational memory.) A project that installs ddflow and stops has the others wired to nothing — and because an agent gate passes on an assertion, that gap is invisible in exactly the way the rest of this design exists to prevent.

So the gap is named:

$ ddflow companions
Companion tools

  [x] context7   Current library documentation
       gates: research, standards
       registered for: claude, cursor
  [x] roborev    Automated second-opinion code review
       gates: standards, bug_hunt, dedupe
       installed (roborev 0.9.1). A cli tool — the agent shells out to it, so
       there is nothing to register.
  [+] codeguide  Language and framework coding standards
       gates: standards
       installed (…) but no agent is configured to launch it.
       -> ddflow companions add --id codeguide
  [ ] sequential Structured step-by-step reasoning
       gates: research, rubber_duck, bug_hunt
       not here: `npx --no-install @modelcontextprotocol/server-sequential-thinking` exited 1
       -> ask the operator, then: npx -y @modelcontextprotocol/server-sequential-thinking

Gates in this project's task pipeline with no companion behind them:
  implement, rubber_duck, critic, unit_tests, bug_hunt, dedupe, merge

A companion counts as registered only when an entry under its id can actually launch something: a bare or junk table with its name does not hide a real launch registered elsewhere, and does not pass for one. A non-table mcp_servers is refused with the parser's reason, and companions add appends TOML only when the result still parses. A stale entry under the id is refreshed in .codex/config.toml too, as in .mcp.json: the old [mcp_servers.<id>] table (and its sub-tables) is cut out and the registry's launch appended. A table that cannot be cut out safely (an inline or dotted form) is refused with "replace it by hand", and a launch that already runs under another name is not doubled.

Three states, reported separately because the remedies differ: registered, installed but not wired up (one command away), not installed (with the command and the URL). ddflow adopt prints the same summary, so the gap is visible at adoption rather than discovered six tasks later. Exit 2 when a default companion is missing — "no data", never collapsed into "no problem".

A server is found by its launch as well as its name. A project that registered codeguide-mcp as coding-guides before ddflow knew it reads as registered for: claude (as `coding-guides`), and companions add leaves it alone instead of writing a second copy under the id. "The same launch" is the same command (by basename) with the companion's arguments in order — only flags, and the value of a flag known to take one (-e TOKEN) may sit between them, the server's own arguments may follow, and a version tag on an npm package (@latest, @1.2.3) is ignored. Another image, package or launcher is another server, and so is reordered arguments.

rules never shows as uncovered: ddflow brief serves it from ddflow's own lessons and operational memory (ddflow memory add|list|forget), which is the job the memory companions below were once recommended for.

Servers and command-line tools are different things, and the registry says which: kind = "mcp" is registrable into an agent's config, kind = "cli" is a tool the agent shells out to. OptMem and pre-commit are examples — real tools with no MCP mode, so companions add refuses them and says why instead of writing a launch entry that would fail its first handshake. A cli companion counts toward its gate's coverage once it is installed; registered is a state it cannot reach.

Your stack needs servers this registry cannot know about. ddflow prompts show research-companions walks an agent from the pipeline's uncovered gates, through the repository's actual manifests, to candidates checked against their primary sources — provenance, maintenance, what they execute, what credential they want — and produces [[companion]] blocks you can read and delete. It proposes; you install. A rejection is part of its report, so the next session does not re-research it.

Registering is previewable. ddflow companions add --dry-run (and ddflow_companions_add with dry_run=true) reports the exact config entry it would write and writes nothing — not the file, not even its parent directory. The handshake tells an agent to dry-run first and show the operator the actual entry rather than a description of it, because registering changes which processes their agent launches. The preview is asserted to match what the real write produces; a preview that drifts from the write is worse than none, since the operator has now signed off on it.

ddflow never installs anything itself — running an install command on someone's machine is the operator's decision. What it does instead is instruct the agent to ask: the MCP instruction block lists each missing companion with the gates it serves and the exact command that would install it, and tells the agent to put that to the operator early, install it if they agree, and record the affected gates unavailable if they decline. Never on its own word.

companions add also refuses to register a server that is not present: that writes a launch command which fails mid-task, at the moment a gate told the agent to reach for it. Detection is read-only and bounded — and when it has not run, the state is reported as unknown, not as absent. ddflow companions probes; the MCP handshake does not, because making an agent wait on npx before it can do anything is the wrong trade.

Adding a fifth is a TOML block in .ddflow/companions.toml, not a patch:

[[companion]]
id      = "my-linter"
title   = "House linter"
gates   = ["standards"]
detect  = ["my-linter", "--version"]
command = "my-linter"
args    = ["mcp"]
install = "cargo install my-linter"

Wiring it into your agent

ddflow adopt --agents claude,cursor,codex writes everything below. This table is what it writes, so you can check it or do it by hand.

7 more are supported with no MCP file to write — a verified absence, not an unresearched gap. adopt writes the delta doc and the AGENTS.md block and names the one manual step. Inventing a path would be worse: ddflow would write a file the agent never reads, and you would believe it was wired up.

One set of rules, every agent

AGENTS.md is the cross-agent convention and most of the 22 read it. Seven do not read it first, or at all, so adopt writes the same managed block into their own surface too:

The rules are inlined, not pointed at. A one-line "see AGENTS.md" stub was the obvious design, and ddflow's own notes had already refuted it: a link is only followed if the agent chooses to follow it. A rule that binds only when the model feels like opening a file is not an enforced rule.

That means several copies of one text, and the answer is that a check owns them: one generator, a managed DDFLOW:BEGIN/END block in each, and ddflow doctor comparing every copy against the generator. Five kinds of break are reported and each fails doctor:

$ ddflow doctor
note:    QWEN.md's ddflow section is from an older version and has drifted
note:    .clinerules/ddflow.md exists but its ddflow section was removed
PROBLEM: .goosehints does not exist — the agent has no project rules at all
PROBLEM: .cursor/rules/ddflow.mdc exists but does not bind: `alwaysApply` is not
         true, so the agent may never load it
PROBLEM: .aider.conf.yml exists but does not bind: it does not list `AGENTS.md`
         under `read:`, and Aider loads no instruction file it was not told to load

Files the project already owns — QWEN.md, replit.md, .goosehints — get a block merged into them; your own content stays. Aider's read: list is extended, not replaced. Adopting is idempotent: re-running never appends a second block.

Every other agent reads AGENTS.md directly, which is the point of it being canonical.

Cursor gets its own rules file because its precedence puts project rules above AGENTS.md — writing only AGENTS.md there would be writing to a file the agent outranks. adopt merges into these files rather than overwriting: they hold your other servers and your other rules, and a tool that stomps them is a tool you run once.

The MCP entry is one line in any of them:

{ "mcpServers": { "ddflow": { "command": "uvx", "args": ["ddflow-mcp"] } } }

uvx fetches and runs it in an ephemeral environment on first use — no clone, no PYTHONPATH, no install step to forget. Prefer Docker? docker run -i --rm -v "$PWD:/repo" ghcr.io/delian/ddflow-mcp, which needs the repo bind-mounted because ddflow operates on your actual git checkout.

Standalone, with no MCP at all, is a first-class mode rather than a fallback. Add to AGENTS.md / CLAUDE.md:

This project's work is a queue managed by ddflow. Before doing anything, run
`ddflow brief`. Claim before you edit (`ddflow claim <id>`), satisfy every gate
(`ddflow gate status <id>`), then `ddflow merge` and `ddflow complete`.
Never pass a gate you did not perform — record `unavailable` with the reason instead.

That is the whole integration. An agent with nothing but a shell can drive the entire workflow, which is why MCP is a convenience layer here and never a requirement.

From plan mode to the queue

Agents plan well and forget reliably. A plan that lives in a chat transcript is gone at the next session; a plan in the queue survives, fans out to parallel agents, and carries its own gates.

Tell the agent, at the end of planning:

Put that plan in ddflow before you build any of it. One phase for the whole plan, one
task per independently-shippable step. Declare each task's globs — the files it will
write — and its needs, the tasks that must finish first. Then show me `ddflow next`.

What the agent does with that:

$ ddflow phase add P3 --title "Rate limiting"
$ ddflow task add P3.T1 --phase P3 --globs 'limiter/**'      --title "token bucket"
$ ddflow task add P3.T2 --phase P3 --globs 'api/middleware/**' \
      --needs P3.T1 --title "wire it into the request path"
$ ddflow task add P3.T3 --phase P3 --globs 'docs/**' --needs P3.T2 --title "document it"
$ ddflow next
Ready (1 ready, 0 running, 2 blocked):
  P3.T1  token bucket
      writes: limiter/**
  (blocked) P3.T2: deps — P3.T1 is open
  (blocked) P3.T3: deps — P3.T2 is open

The two fields that do the work are --globs and --needs. Globs are how two agents are stopped from editing the same file: claim refuses an item whose writes overlap one already held, and names what to take instead. Needs are how ordering is enforced without anyone remembering it. A plan whose tasks declare neither is a list, not a queue — it will look parallel and then two agents will fight over one file.

ddflow split <id> --into a,b,c exists for when a task turns out to be three, which is the normal case rather than a failure of planning.

When a companion is missing

ddflow imposes the order and demands the evidence. It does not perform the judgement inside most gates — that is what the companion tools are for. So the honest question is what happens when one is absent, and the answer is deliberately never "the gate passes".

The rule, and it is enforced: a gate whose tool could not run is recorded unavailable with the reason, never passed. ddflow complete reports those as a coverage gap on the completion event, so a finished item never silently implies that a check happened. Set [gates].unavailable_is_failure = true and a gap blocks completion outright.

ddflow companions reports four states, and the difference between the last two is the whole point: registered, installed but not wired up (one command away), missing (with the install command and the URL), and not checked — because the MCP handshake does not probe, and "nobody looked" must never render as "not there".

ddflow never installs anything. Detection is read-only and the report is advice. ddflow companions add --dry-run shows the exact config entry it would write, so an agent can show you the change before making it.

Adopting a project that already has history

A queue that starts empty tells the next agent "nothing is in flight" about a repository with three branches in flight and forty open items in a todo file — and the agent believes it, because the tool said so. That is worse than having no tool at all.

$ ddflow import                      # looks; writes nothing
What this project already has (nothing written yet):

  314 phase(s):
    [ ] 142.A     the scaling-law advisor is wrong (P0; CONFIRMED)   docs/todo.md:26517
  1170 task(s):
  442 lesson(s):
  ...
  47 memory(s):
    [ ] M-0002    Hardware: 8x H200 GPUs on this box, usually idle.  .agent_memory/LOG.txt:3

  NOTE: 3631 already-ticked task(s) were NOT imported. They are history, not a queue.
  NOTE: 32 phase heading(s) say the work is finished while their checkboxes are still
        unticked: 99 (4 open), 103 (3 open), ... Ask the operator which is stale.

$ ddflow import --apply              # writes them, each recording its source line

Seven sources, all optional, all in the places projects actually keep them:

What it will and will not decide for you

Mechanical, and verifiable: a ticked checkbox is a fact, a ## heading is a section, a branch with unmerged commits is work. The id in ### 142.A — … or - [ ] **WFOPT.4.6** — … is read, not invented, so the imported queue uses the ids the project has been writing in commit trailers for months.

Judgement, and yours: which open items are actually live, what each task writes, what depends on what. The /import-existing-project prompt walks an agent through that with the operator. It is not automatable, and a confident guess produces a wrong queue the scheduler then hands out.

Four guard rails, each of which exists because the alternative is silent:

• Dry run by default. --apply writes. Looking is free and never a side effect.
• Finished work stays out — it is history, not a queue — except a completed item that open work depends on, which comes along as done so the open item is not stranded on an id the queue has never heard of. A needed phase is judged over every task under it, so a ticked task pulled in as a dependency does not keep the phase open, and a plain import completes a finished phase that open work needs (re-running repairs one left empty).
• [importer] max_tasks (default 200) refuses a whole history. An import writes events into a log that is committed to git; one real repository yielded 4,799 checkboxes. Over the cap it proposes none and says so — the phases are withheld with them, because a queue of empty phases is not a smaller import, it is a misleading one.
• Idempotent. Ids derive from the source, so re-running after you edit the todo adds what is new and leaves the rest alone. A second run over an unchanged project exits 2.

What an unticked box means, and where things live

An open box is not always work. The import reads the project's own dispositions — the vocabulary and positions are those of the picker ddflow was extracted from, and agree with it on 1,166 of that repository's 1,170 open boxes. Of the four, two are the picker's own false positives ("cells run / skipped" in plain prose) and two are recorded non-findings with an unclosed "(… refuted it" aside, which ddflow closes:

[importer] archive_globs names plan files that are history until a section is named (a 20,000-line legacy docs/todo.md): their open boxes import blocked, and ddflow unblock <phase> releases a whole section at once.

Each source family's location is a knob — todo_globs, lesson_globs, lesson_summary_globs, decision_globs, research_globs, journal_globs, memory_globs — and a set knob replaces the defaults, because the same filename means opposite things in different projects (docs/LOG.md is one repository's whole journal and another's generated index of it; an Index section is never imported).

Lessons are split at the level they actually live at: ### L100. … entries grouped under ## <date> headings import one per lesson with their own ids, so [L147] cross-references still resolve — and so do hyphenated ids (L-12, R-7), kept as ids in headings and citations alike. A lesson's **Compressed:** paragraph becomes its summary; a hand-written lessons-summary.md bullet that cites exactly one lesson becomes that lesson's summary, and every other bullet becomes a consolidated lesson tagged summary. A GENERATED summary file is skipped. ddflow render writes them all back out as docs/ddflow/LESSONS-SUMMARY.md (also ddflow://lessons-summary).

An import adds nothing twice, and says what repeats. Lessons, decisions and research are weighed against every record already in the log (and a summary bullet against the lessons of the same import) with the same similarity engine and [dedupe] thresholds an add uses. Text identical to a held record is left out and counted; a record scoring at least [dedupe].ask_threshold (0.55, with min_words content words) is left out and LISTED in the plan's notes as lesson LS-git-rebase ~ L12 (0.74) docs/lessons-summary.md:9 -- the candidate id and score -- so the operator or onboarding agent decides: nothing is written for it, and filing one anyway is ddflow lesson add (or decision add / research add) once the existing record is judged not to say it. No score separates a duplicate from a related record, which is why it is reported rather than decided. On a real project's lessons-summary, 68 of 86 bullets that restate a corpus lesson were reported this way. Tasks and phases are not checked (they carry dependencies); [dedupe].on_match = "warn" reports the same list but imports them anyway, and "off" turns the check off. Under the shipped default, ask, near-duplicates are withheld. Re-running over the same files adds nothing.

Verifying an import, at any time

The import's weak spot was never the parsing. It is everything after --apply: 1,170 tasks arrived in the real-corpus run, and the workflow prompt tells an agent to give each one globs and declare its dependencies. Nothing checked whether that ever happened — and an imported queue nobody finished misrepresents the project exactly as an empty one does, believed harder because a tool produced it.

$ ddflow import --verify
Imported between 2026-09-25 and 2026-09-25:

       4 decision(s)
    1727 journal(s)
     442 lesson(s)
      47 memory(s)
     314 phase(s)
      72 research(s)
    1170 task(s)

Left to decide or fix:
  - 1078 imported task(s) declare no globs, so the conflict detector cannot protect
    them and two agents can be handed the same file: OPIK.1b, OPIK.2, ...
  - 27 phase(s) say the work is finished while a task under them is still open: 99,
    103, 115.D.2, ... Ask the operator which is stale before anyone claims from them.

Three answers, three exit codes, because collapsing them loses the one that matters:

It reports status (what is imported, per kind, and when), whether it is still true (what a re-run would add, which sources yielded nothing, which source files have since vanished), and whether anyone finished it (tasks with no globs; phases whose heading claims SHIPPED over an open task).

It deliberately does not repeat ddflow doctor, which already reports unresolved dependencies, duplicate globs and cycles. Two commands reporting one defect in different words is how an operator learns to read neither.

Provenance is a field, not prose. Item.source is docs/todo.md:41; the body still says "Imported from docs/todo.md:41." for a human reading ddflow show. Answering "which items came from the import" by regexing that sentence would mean the day someone rewords it, the count silently becomes zero and the verification passes.

The connection handshake follows through. The offer to import stops once the queue has anything in it — but if imported work is still missing globs, or a phase still claims SHIPPED over open tasks, the MCP instructions say so and tell the agent to run ddflow_import_verify before handing any of it out. That check is computed from the already-folded queue, so it costs nothing; the source re-scan (~0.65 s) stays out of every session start and happens only when someone asks for it.

Re-running is a first-class path. /import-existing-project opens by checking what is already imported and switches to finishing and refreshing rather than repeating — fix the globs it names, ask the operator about the SHIPPED drift, re-run ddflow import for sections added since.

What it reports rather than fixes

Three kinds of drift it can see and must not resolve on its own, because either answer could be the wrong one:

• A phase heading that says SHIPPED over unticked checkboxes (32 of them in the repository this was measured against). One-sided risk: if the heading is right, the queue is about to hand out work that is already done.
• A dependency on an id nothing produced. Kept and treated as unmet — deliberately, so a typo surfaces as blocked work rather than as work that starts early — but named, because "never offered" otherwise looks exactly like "nobody has got to it yet".
• A file that matched a source pattern and yielded nothing, which usually means an unusual format rather than an empty file.

The model: phases, tasks, dependencies, globs

Plan ──► Phase ──► Task

A phase is a unit of review: its own research, its own whole-suite test pass, its own live smoke run, merged as one coherent feature. A task is a unit of execution: one agent, one worktree, one pipeline, one merge. Both carry needs (dependencies, which may cross phases) and globs (the files they will write).

ddflow phase add P2 --title "Billing" --needs P1
ddflow task add P2.T1 --phase P2 --title "invoice model"  --globs "src/billing/invoice.py"
ddflow task add P2.T2 --phase P2 --title "tax rules"      --globs "src/billing/tax.py"
ddflow task add P2.T3 --phase P2 --title "checkout wiring" --needs "P2.T1,P2.T2" \
                                                            --globs "src/checkout/*"

Declare globs. They are what lets two agents work at once safely. A task with no declared globs is a task the conflict detector cannot protect.

Dependencies are inherited. A phase is never claimed — only its tasks are — so P2 needs P1 has to govern everything inside P2, or it governs nothing that anyone picks up. The readiness rule therefore consults an item's ancestors as well as itself:

$ ddflow next
Ready (1 ready, 0 running, 1 blocked):
  P1.T1  money
  (blocked) P2.T1: deps — phase P1 has 3 open task(s) (inherited from P2)

The refusal names where the dependency came from, because an operator told only "P2.T1 needs P1" goes looking for a declaration that is not written there. The one dependency not inherited is one pointing into your own subtree: an umbrella that declares a dependency on its own child would otherwise make the child wait for itself, turning a plan typo into a permanent hang.

ddflow claim asks the same predicate ddflow next does. They used to disagree — next withheld a task on its dependencies and claim handed out a worktree for it a second later — so an agent picking work by id rather than by asking bypassed the dependency graph entirely.

There is deliberately no third level of kind: a sub-task is a task whose parent is a task, so depth is unlimited while the rules stay one set.

Model-tier hint (advisory)

Tag a task tier:fast, tier:balanced or tier:deep (ddflow task add ... --tags tier:fast) to say what kind of model suits it: a cheap one for mechanical bulk work, a balanced one for implementation, a top-tier one for architecture trade-offs. ddflow next (CLI text, --json as a tier field, and the bounded ddflow_next body) and ddflow brief (the item's header and the ready list) show it; an untagged item shows nothing extra. A harness that dispatches a subagent may map it to its model choice (see the driver, docs/ddflow/drivers/implement-phase.md). It is advice only: it never affects scheduling, gates or reviewer independence, and an unknown value such as tier:foo is ignored and reported by ddflow doctor as a note, not an error.

Work that changes shape while you do it

Tasks can be added at any time, including while their parent is being worked — mid-task discovery is the normal case, not an exception, and a queue that cannot absorb it pushes the work into someone's head.

Sub-tasks are just tasks whose parent is a task. Not a separate concept with its own rules: a sub-task declares its own globs, carries its own dependencies, is claimed by its own agent, and runs in parallel with its siblings when nothing links them — exactly like any other task.

ddflow task add P1.T1a --parent P1.T1 --globs "src/parse.py"
ddflow split P1.T1 --into "P1.T1a=parse input" --into "P1.T1b=write records"

split works in place: the original keeps its id, its lease history and everything recorded against it, and becomes an umbrella that completes when its children do. Closing it and opening two new ones instead would lose the thread between what was planned and what happened — which is exactly what ddflow replay needs.

An umbrella is never offered as ready (its children are), and cannot complete while any descendant at any depth is unfinished. An abandoned child counts as settled, so a sub-task you decide against does not hold its parent open forever.

Becoming an umbrella releases the lease, however you get there — by split, or by adding the first sub-task to a task you are already working. An umbrella holding a live claim on globs that overlap every child's means a second agent cannot take one of those children, and crash recovery points at a worktree where nothing further will happen. split already did this; task add --parent did not, which is the shape of bug worth naming: one transition, two ways in, guarded on one.

Architectural decisions

The code shows what was built and never why, nor what was rejected on the way. So decisions are recorded as events, and reach the person writing the code:

ddflow decision add --title "Storage is SQLite with WAL" \
  --decision "One file, WAL mode, BEGIN IMMEDIATE for writes." \
  --context "Three call sites were each opening their own connection." \
  --alternatives "Postgres — rejected: no server allowed in this deployment." \
  --globs "src/storage/*" --by operator

--globs is what makes a decision consulted rather than merely filed. ddflow brief and ddflow decision applicable <item> surface the decisions governing an item's declared files automatically — the agent does not have to suspect they exist.

Decisions are never edited or deleted. A reversal is a new decision naming the old one (--supersedes), so the history of how the architecture got here survives, and a superseded decision is shown with a pointer to its replacement rather than silently withheld.

Recall — "have we been here before?"

ddflow recall "how should durations be represented"

One search across everything the project remembers: architectural decisions, lessons, operational memories, research verdicts, past bugs, similar tasks, and the operator's own earlier prompts. Results are labelled by kind, because a binding decision, a transferable lesson and a prompt from three weeks ago should change what you do in different ways.

It exists so the operator does not have to say the same thing twice and the agent does not have to learn the same thing twice. Both failures are invisible in the moment and obvious in the log.

Which existing record is this new one like? The similarity engine answers that without an LLM: a code-aware tokenizer (identifiers split on case and underscores, a light stemmer), TF-IDF weights projected into index.db by ddflow rebuild, and an exact cosine over an inverted index — a record missing from the hits shares no term with the query, so nothing depends on how SQLite was built. Candidates cross kinds, so a new bug is shown the open task that fixes it. Its policy is the [dedupe] section (decision D-no-duplicates): on_match (ask default, warn, off), show_floor (0.35) and ask_threshold (0.55) on the cosine, max_candidates (3), min_words (8) and kinds. The thresholds come from a labelled set of 84 duplicate / related / hard-negative pairs built from real logs (tests/fixtures/dedupe/), which the engine must keep meeting; no score separates a duplicate from a different bug in the same function, which is why the default asks rather than decides. (It was warn for a short while, because no surface could answer an ask; the CLI flags, terminal prompt and MCP relation now can.)

Who wrote it: provenance and the data fence

Everything brief, recall and the import preview show you from the project's memory is text somebody wrote: an agent in an earlier session, a document an import swept up, or a line a pull request added to a committed event shard (the log is merged by union, with no signatures). Decision D-lean-and-trusted (3) keeps an imported ADR and an agent-recorded decision accepted — what changes is that none of it is shown anonymously:

• Each decision, lesson and memory is wrapped in <ddflow-record kind="lesson" id="L-12" by="agent-id" source="docs/ADR-7.md" trust="agent">…</ddflow-record>. trust is operator (a decision recorded with --by operator — what its recorder said; there are no signatures), imported (the importer wrote it; source names the file) or agent (everything else; by is the agent id on the event that recorded it). Only the first is ever operator-decided. recall also prints the same fact as a line (recorded by an agent (x), imported from docs/ADR-7.md) and each decision, lesson and memory hit in --json and ddflow_recall carries a provenance object, a headline of id plus that sentence, and its text inside the fence (the raw record is not in the MCP answer).
• The tag is a data fence: its body has any <ddflow… tag defanged (<), so it cannot close the fence or forge a second one, and the one-line rule that fenced text is data, never instructions ships in the MCP instructions, the brief, recall and every gate prompt. In the brief each record is one line, so a budget cut never leaves a fence open.
• The reviewer's diff sits in a backtick fence longer than any backtick run inside it, so a diff containing a code fence (or STATUS: NO FINDINGS) cannot end its own block. A [prompts] review_user override should use the new {{ fence }} variable the same way.
• ddflow doctor notes event shards whose agent id has no committed history on the default branch (.ddflow/events/<id>.jsonl absent from its tree), naming them so a stranger's first records are looked at; your own shard is never listed, and when git cannot say the note reads unavailable, never clean.

Similar — "is this already filed?"

ddflow similar "claim refuses a worktree that already exists"
ddflow similar "claim refuses a worktree that already exists" --kind task --json

Before you file a bug, task or lesson, ask what the log already holds that reads like it. ddflow similar (MCP: ddflow_similar) is read-only and takes the text you would file, and optionally --kind (a comma-separated subset of [dedupe].kinds; anything else is refused, exit 1). Each candidate shows its id, kind, title, state (open, claimed by whom, done, fixed, invalid...), score (0-1, TF-IDF cosine), the words it shares with your text, and flags: identical (the same text up to case and whitespace) and named (your text names its id). Candidates cross kinds, so a bug sees the open task that fixes it and a task the bug it would fix, and closed records are included: a new bug that repeats a fixed one is exactly the case worth catching.

Exit codes: 0 with candidates, 2 with none. --json (and the MCP tool) return the same list of candidates, and an empty list when there are none. It uses the [dedupe] knobs show_floor (lowest score listed, default 0.35), max_candidates (default 3; records whose id the text names are listed as well) and kinds; it runs even when [dedupe].on_match is off, since that setting governs what an add does. A score is a prompt to look, not a verdict: two bugs in one function score high and are different.

The check every add runs

Every add — task, phase, bug, lesson, decision, research and memory; never a session prompt or note — runs the same check against the log before it writes, with the [dedupe] knobs above. What it does depends on how close the match is:

• A match — a candidate scoring at least ask_threshold (0.55) when the new text has at least min_words (8) content words, or text that names an existing record's id — is refused (exit 3, refused: possible duplicate). Nothing is written. The refusal carries candidates (id, kind, title, state, score, the words shared) and options (new, then extends / duplicate_of / related for each candidate), and its reason lists them in prose. Over MCP the refusal leads the reply as {"refusal": {...}}. A bug filed against the task that fixes it (--item T) already answers that candidate and is not asked about it.
• An answer lifts the refusal. new: it is a different record, filed as asked. extends X / duplicate_of X: it is the same thing. While X is open and unclaimed (an open task or phase, an unfixed bug, a lesson or decision not superseded, research, a live memory) the text is appended to X as a record.extended — verbatim, with who, when and the score — and no new id is made; a bug's own summary is never replaced. If X is claimed, running, done, fixed or superseded, a new record is filed linked to X (extends / duplicate_of on its add event), so nothing is lost, and the result names X and its state so the holder can be told. related X files a new record linked both ways (a link.recorded on X). The API takes the answer as one answer argument (api.DedupeAnswer("extends", "B5d98a4da0a"); DedupeAnswer.parse("related B1")) carried on task_add, phase_add, bug_found, memory_add and the lesson, decision and research drafts. The CLI and MCP surfaces are below.
• Answering a refusal, on every surface.
  • CLI flags on every add command (task add, phase add, bug found, lesson add, decision add, research add, memory add): --new, --extends ID, --duplicate-of ID, --related ID and --check; at most one of the five (argparse refuses two, and over MCP relation with check_only is a failure). --check is a dry run: it writes nothing, prints the candidates and whether the add would be refused, and exits 0 with candidates or 2 with none (--json prints candidates, options, would_ask, and for an exact copy would_extend -- the open record it would be added to -- or would_link -- the claimed or closed record it would be filed beside). An id that already exists keeps its own rule and is not checked (--check says so, exit 2 -- except a task or phase, whose existing id is the queue's ordinary already exists refusal, exit 3, exactly as the add itself would answer); an index that cannot be read is exit 1, never "none". ddflow similar "<text>" asks the same question before you have an id or a command to run.
  • On a terminal (stdin and stdout are both terminals, no --json, no answer flag) a refused add asks instead of failing: it lists the candidates numbered and prompts [n]ew / [e]xtends # / [d]uplicate of # / [r]elated # / [a]bort. e 1 and r T-old both work (the number or the id; the number alone is enough when there is one candidate). Anything else asks again; abort, or end of input, files nothing and exits 3.
  • Without a terminal (a script, an agent's shell, a pipe) it exits 3 with refused: possible duplicate, the candidates, and the same command again with each answer appended -- --new, --extends TOP, --duplicate-of TOP, --related TOP -- ready to paste. --json prints the same refusal as candidates plus options (the reason is still on stderr). Nothing is written. An add that goes ahead anyway -- warn, or a match below the asking threshold -- prints It reads like: and the candidates on stderr before its usual line.
  • MCP: every add tool (ddflow_task_add, ddflow_phase_add, ddflow_bug_found, ddflow_lesson_add, ddflow_decision_add, ddflow_research_add, ddflow_memory_add) takes relation -- new, extends:ID, duplicate_of:ID or related:ID -- and check_only (the dry run). The refusal leads the reply as {"refusal": {...}} with candidates and options beside it; answer it by calling the tool again with relation. An answer that lands on an existing record returns extended (the record that received the text), extended_kind and relation in place of a new id.
  • Agents: run ddflow similar before filing; prefer extending an open, unclaimed record; a claimed or closed one gets a new record linked to it (extends does that on its own).
• Identical text (up to case and whitespace) as an existing record of the same kind is recorded as a duplicate of it without asking, under the same open-or-linked rule.
• Below the threshold, at or above show_floor (0.35): the add goes through and its result lists the candidates.
• Every answer is recorded on the add event (dedupe: the answer, the score, the candidates shown), new included, and an automatic merge is marked auto.

Seeing what was added. ddflow show X (an item or a bug id; --json carries the same data as additions, links and linked_from) lists the additions on X verbatim with who, when and the score; the links X makes (links); the records that link to X (linked_from) — found by scanning every record's links for X, because a new record filed extends / duplicate_of X holds the link itself and only related also writes a back-link, which is listed once; and, for a bug, the task that fixes it. The holder of a claimed item is told what arrived: ddflow brief for it leads with N new reports on your item since you claimed (additions made, and records linked to it, since its lease was acquired), quoting each (the first five, clipped; show has the rest) inside the brief's token budget, and ddflow heartbeat and ddflow gate status carry the count in one line. MCP ddflow_show and ddflow_brief return the same data.

[dedupe].on_match sets the policy: ask (default) as above, warn never refuses or merges — it lists the candidates and records the add as new — and off skips the check entirely. Adding an id that already exists keeps the refusal or merge it always had. An id ddflow generates (no --id) is time-salted, so the same text filed twice gets two different ids; only the duplicate check, never the id, catches a re-filing: an exact copy of an open record is merged into it as an extension, of a closed one filed as a new record linked to it, both without asking under the default ask (warn and off leave two unlinked records). The check reads the log before the add writes, so it is advisory across agents: two adds of the same text racing in different clones can both pass, and a later similar will show the pair. The check is also a function, api.dedupe_check_add, taking a config that may have on_match replaced (api._dedupe.with_check), so a bulk caller need not go through the add paths. The importer applies the same engine and [dedupe] thresholds to a batch with its own reconciliation (see the import section), not through this function.

Operational memory

ddflow memory add "8x H200 on this box; check nvidia-smi before a GPU test" --tags gpu
ddflow memory list                       # newest first; --query to rank, --all for forgotten
ddflow memory forget M-0003 --reason "the box was upgraded"

One fact about this machine, repository or working state — not a rule (lesson), not what happened (session note), not how the software is built (decision). Capped at [memory] max_chars (280) and refused, not truncated, when longer. The newest [memory] brief_items appear in every ddflow brief, right after the binding decisions, and recall searches them — the job an OptMem store beside the repository used to do, now in the log, so every worktree sees a memory the moment it is written. A memory that stopped being true is forgotten with a reason, never deleted: "we thought X until Y" is what stops the next agent re-learning X. An OptMem LOG.txt imports as memories dated when they became true. The log is committed: never put a secret in one.

Resources and long-running jobs

Globs keep two agents out of one file. Work that runs on something — GPUs, a model server, a shared fleet — declares that too, and the same refusal applies:

ddflow update TRAIN.3 --resources gpu:6            # or **Resources:** gpu:6 in the plan
ddflow claim TRAIN.3                                 # refused (exit 3) if it does not fit

[schedule] resources = ["gpu=8", "vllm-fleet=1"] sets capacities; a resource named nowhere is exclusive. Every live claim counts, the claimant's own included — one agent starting two 8-GPU runs overcommits the box just the same — and next withholds what does not fit, saying who holds what.

The run itself is a job:

ddflow job run TRAIN.3 "uv run main.py train -c cfg.toml"   # detached; survives you
ddflow job list        # running | exited N (from its log) | gone (killed) | elsewhere
ddflow job end J3f2 --note "loss 0.12, ckpt in out/"         # refused while it runs

A launched job runs in the item's worktree, in a session of its own (it outlives the agent, the MCP server and a restarted remote-control service), and appends its exit code to its log so a run nobody watched still says how it ended. Liveness is computed, not stored: a zombie is not alive, and a reused pid is caught by the process start time. ddflow job add --pid registers a process started some other way. Every brief lists jobs not yet recorded as ended — "WAIT, do not start it again" for a running one.

Dependencies on another repository

[schedule]
repos = ["run_nemo_run=../run_nemo_run"]

ddflow update GEN.4 --needs run_nemo_run:132.D then waits for item 132.D there to be done. ddflow external sync (and every session-start hook) reads the sibling's log — never writing it — and records what it observed in this one, only when it changed. Readiness is decided from that dated fact, so the fold stays pure and "why was this started?" is answerable later. An unobserved external dependency is unmet; one naming a repository that is not configured is a doctor problem, since it can never be met.

Status, progress, and loops

ddflow status      # what is done, in flight, ready, blocked — one answer
ddflow progress    # attempts, hours held, gate runs, commits, per item
ddflow loops       # circular references and runtime loops (exit 2 = none)

progress counts a landing once (commits holds distinct shas). ddflow show <id> takes a bug id too: its state, where it was found, the fix task(s) — a task that says "fixes bug X" is a fix, a mention is not — the regression tests it was closed with, and an invalid closure a later fix superseded. ddflow --json board prints the board as JSON (a task under no phase is listed in an Unphased section of the board, and under unphased in the JSON); --agent, --repo and --json are accepted after the subcommand as well as before it.

next --phase, brief --phase and board --phase refuse an id that is not an item and next names the phases that start with it; board --phase given a task id refuses, naming that task's phase. ddflow status over MCP is bounded: the counts are exact, but each long list is cut to the 25 most recent (the first 25 in scheduler order for the others), with a truncated note naming the real lengths; the CLI, and ddflow --json status, list everything. The other large MCP reads are bounded the same way, and a cut is never silent: ddflow_next lists the ready items whole and the first 10 blocked ones, with a truncated field giving the exact blocked total and a count per reason; ddflow_show leaves out each gate record's tree ids and merge bookkeeping and cuts any gate or triage string past 160 characters, with truncated naming what went; ddflow_progress (most effort first) and ddflow_decision_list (newest; no context or alternatives, decision text clipped to 400 characters, ddflow_decision_show has one whole) return 25 rows, limit raises it and limit=0 is all, and a second content block states the cut; ddflow_recall drops each hit's raw record and keeps hits within max_chars (default 4000, counted as the JSON returned), one per kind in turn, and says so when it cut any. MCP bodies are compact JSON. --json on the CLI is the whole, indented body in every case. On this repository that took next from 38 KB to 3.9 KB, show of a finished task from 8.9 KB to 4.1 KB, recall from 51 KB to 4.1 KB, decision_list from 54 KB to 18 KB and progress from 158 KB to 8.4 KB. doctor and status also say when the log holds events from a newer ddflow than this checkout runs — they were skipped, so the numbers are computed without them — and the remedy is to merge main or run the newer ddflow.

brief says whose work it is. Run under an identity, its ## Current is the agent's own most recent lease (several are listed, with --item <id> for another); an agent holding nothing sees ## Suggested next — the queue's top ready item, to claim first — because calling it Current told an agent it was working on something it never claimed. complete likewise defaults the author model from the agent's open session start.

Dependency cycles are the easy case. The expensive ones are runtime loops, where the graph is perfectly acyclic and the work still never finishes:

Every threshold is a [loops] knob, and on_detect = "block" makes ddflow claim refuse an item that is already looping — a warning is read by a human later, a refused claim is read by the agent now. repeated_failure also shows as a line in the item's ddflow brief, and under block refuses a gate run of that gate until the work has changed (the refusal names the evidence; a different tree lifts it). max_repeated_failures = 0 turns it off.

The task pipeline

Ten gates, in order, configurable per project:

Four things are enforced rather than requested:

Silence is not a pass. Every gate in the pipeline must carry some outcome before an item completes — passed, failed, unavailable, partial, or an explicit ddflow gate skip <id> <gate> --reason "...". Without this, gates.required held only implement, unit_tests and merge, so six of the ten steps could be omitted with no trace at all. gates.require_outcome = false makes the pipeline advisory again; gates.enforce_order ("warn" by default, or "block") reports a gate recorded before an earlier one has run, because a rubber-duck review recorded before implement reviewed an empty diff.

UNAVAILABLE is never a pass. A reviewer whose endpoint was down approved nothing; a linter that is not installed found nothing. Each gets its own outcome and shows as a coverage gap. (The inverse matters too: this codebase's first version classified a missing binary — shell exit 127 — as failed, so an uninstalled linter looked like a linter reporting problems. Fixed, with a mutation-verified regression test.)

Evidence or it did not happen. Gates in gates.evidence_required reject a bare pass; they want the command, its exit code and its output digest.

And evidence says WHICH tree and HOW MUCH. Every gate that produces an OUTCOME — command gates, and agent gates recorded with gate record — carries a tree_sha: a fingerprint of the working tree it ran against, covering committed state, uncommitted changes to tracked files, and the content of untracked ones (a new module is untracked until its first commit, which is the ordinary state of agent work). ddflow's own .ddflow/ is excluded, or recording a gate's outcome would invalidate the gate that just recorded it. If the tree moves afterwards, complete warns that the pass describes source nobody is shipping — a warning, not a block, because refusing on a comment-sized change is how a check gets switched off. The comparison is of content trees, not commit ids, so an amended message is not a change; once the item has landed it is made against the commit that landed (a merge commit's second parent, which is what the gates ran on — never the primary checkout, whose files are everyone's), and the note names what differs. Evidence that cannot be compared says so rather than passing silently.

Beside it, diff_stat records files, insertions, deletions and untracked count — including the lines in untracked files, because a new module is untracked until its first commit and a task that is entirely new files would otherwise report zero insertions. The fingerprint answers which tree and is opaque; this answers how big, and that is what makes a pass auditable later — a review gate that passed over 4,000 changed lines in two minutes is a different claim from one that passed over 12.

Neither is recorded for a skip (nothing was reviewed, so a magnitude would imply an inspection that did not happen) nor for an unavailable gate that never ran.

Hashing untracked content is capped by MAX_UNTRACKED_HASHED (512). Above it the fingerprint falls back to file names and says so inside the digest, because a check that quietly stopped covering content would go silent for exactly the repositories that need it most.

Reviewer independence is checked, and an unidentified reviewer establishes nothing. Same-family reviewers share the author's blind spots, so their agreement measures shared priors rather than correctness. complete refuses unless one reviewer came from a different pretraining family — and a reviewer whose model is not in [agent].families counts as unknown, never as different. (It used to count as different: gate record defaults the reviewer to the agent id, so a standards gate recorded with no --model arrived as family "host-12345", compared unequal to "anthropic", and satisfied the independence requirement on its own.)

--model on gate record names the reviewer; the author's model is complete --model (or the one declared at session start). Recording the author's own model on a reviewer gate used to overwrite the reviewer's, and complete then judged independence against the author's family. Now a --model in the author's family on rubber_duck/critic is refused (exit 3, naming --reviewer-model); --reviewer-model <m> (MCP reviewer_model) is the explicit spelling for a reviewer that really is of that family. The author is the model the agent declared at session start --model; with none declared there is nothing to compare and the record stands. standards is covered too, since it counts toward independence.

--reviewed-sha ties a review to the commit that merges. roborev review HEAD run from an item's worktree has enqueued the primary checkout's HEAD, so a standards gate was recorded against the wrong commit. Run roborev review <sha> with your branch head and record gate record <id> standards --reviewed-sha <sha>: the full sha is stored in the evidence, an older commit of the item's branch is accepted with a warning, and any other commit (main, another branch, an unknown sha, or a symbolic ref such as HEAD, which means a different commit in each checkout) is refused with exit 3. An item with no branch to compare against records the sha with a note.

ddflow review of an item reviews the branch's commits plus tracked edits in its worktree. Untracked files are not sent to the reviewer; they are listed in the output and in the evidence's diff_source as "untracked, not reviewed" — commit a file to have it reviewed.

A router author is a set of families. Copilot CLI's HydraFusion is chosen like a model but routes each task across models from several providers, so [agent].routers maps its name to the families it draws on, and a reviewer counts as independent only when its family is outside the whole set. HydraFusion ships with an empty set — GitHub publishes no fixed roster — so complete --model hydrafusion refuses, naming the knob, until you list the families your plan routes to:

[agent]
routers = { hydrafusion = ["anthropic", "openai", "google"] }

$ ddflow complete P1.T1 --model claude-opus-5
cannot complete P1.T1 — 1 unmet condition(s):
  - reviewer independence not satisfied: every reviewer (rubber_duck) was family
    'anthropic', the same as the author. Same-family agreement is not independent evidence.

Every unmet condition is listed at once — a refusal that reveals one problem at a time trains an agent to reach for --force.

Proving a gate can fail at all

$ ddflow gate verify T1 unit_tests
  OK   src/calc.py: detected

unit_tests CAN fail: every registered mutation was caught.

A gate that cannot go red is worse than no gate — it reports success on every change, and everyone downstream reads that as evidence. gate verify breaks what the gate guards, using the mutations registered beside it, and requires the gate to notice:

[gate.unit_tests]
command = "python -m pytest -q"
cwd = "repo"
mutations = [ { file = "src/calc.py", old = "return a + b", new = "return a - b" } ]

Four things make it honest rather than ceremonial, and the last one is this feature's own bug, found by a cross-family review of it:

• A mutation that did not apply is a FAILURE, not a skip. If old is absent — or present twice, so the edit is ambiguous — the check fails. Skipping turns "the mutation never happened" into a green run, which reads as the opposite of the truth.
• The source is restored whatever happens, including on exception, or a failed verification leaves the tree broken and the next gate reports the verifier's fault.
• A gate with no registered mutations is reported as unproven. Declaring a check nobody has shown can fail is what this exists to catch. An agent gate says plainly that it has no command to mutate and rests on its evidence contract instead.
• A green baseline is required first. A gate already red for an unrelated reason — one pre-existing failing test, a tool that stopped being installed, a flake — reports failed for every mutation, so every mutation reads as detected and the gate is certified as able to fail when nothing has shown any such thing. The check written to catch the vacuous-pass class contained it. It now runs unmutated first and refuses without a pass.

When the exit code is not the verdict

Some tools say "I could not run" or "I only did part of it" with an exit code, and some exit 0 whatever happened. A command gate can say which:

[gate.critic]
command = "uv run scripts/critic_review.py --dirty -c configs/review_critic.toml"
unavailable_exits = [2, 143]      # endpoint down / SIGTERM: UNAVAILABLE, not failed
partial_exits = [3]               # reviewed part of the diff: PARTIAL
require_output = '^STATUS:'       # exit 0 without it = the tool did not do its job
fail_output = '^\s*- \[(HIGH|MEDIUM)\]'   # exit 0 WITH findings = failed

ddflow gate run also renews the caller's lease every [lease] heartbeat_s while the command runs, so a 25-minute suite does not outlive a 30-minute lease and read as abandoned work.

It keeps the run's whole output in .ddflow/runs/<item>/<gate>-<time>-<pid>.log (the newest ten per item and gate; the directory ignores itself, so nothing is committed) and records its path beside the digest, so the digest can be checked against bytes that still exist. The evidence also carries the suite's own verdict lines (summary: pytest's === 3 failed, 112 passed ===, unittest's FAILED (failures=2)) lifted from the whole output, because a gate that reruns its failures ends on the rerun's "112 passed" and the tail alone would hide the first pass's "115 failed". gate record --output-file keeps the file's path the same way.

The phase pipeline

research → [ task, task, task … ] → unit_tests → bug_hunt → dedupe
         → live_test → corrections → docs → merge

docs runs once per phase, before it merges: every user-visible change the phase made (commands, MCP tools, flags, knobs and defaults, output, setup) is checked against the README and the docs and updated there, with the files changed as its evidence. The per-commit stale-docs check catches a renamed or removed name; this catches the new feature nobody wrote down.

live_test is the one most often skipped and the one most worth keeping: a green unit suite and a working feature are different claims. Run the real thing on a small input and paste what it printed.

Human approval: a gate the agent cannot clear

Every other gate here is satisfied by the agent — it runs a command, or it asserts it did the thinking. That is right for work whose correctness is checkable afterwards, and wrong for a plan: by the time an agent has built the wrong thing, the cost is already paid.

A gate marked human = true is where the operator says yes, build that before the compute is spent.

# .ddflow/gates.toml
[gate.plan_approved]
title  = "Operator approves the plan"
human  = true
prompt = "Show the operator what you intend to build, then ask."

$ ddflow gate run T1 plan_approved
gate 'plan_approved' is a HUMAN-APPROVAL gate. It is not something you can run or
record — it is where the operator decides whether this work should proceed.
  ddflow approve T1 plan_approved
  ddflow approve T1 plan_approved --reject --reason '...'

$ ddflow gate record T1 plan_approved --outcome passed --evidence "looks fine"
'plan_approved' is a human-approval gate: it is cleared by a person, not by an agent
recording that it happened.                                            # exit 3

$ ddflow approve T1 plan_approved --note "read the plan, ship it"
T1.plan_approved approved by delian — read the plan, ship it

A rejection is a first-class outcome, not the absence of an approval: "the operator looked and said no" and "nobody has looked yet" are different states, and an item sitting in the second forever is how a checkpoint becomes a silent stall. --reject requires --reason.

There is deliberately no MCP tool for this, and tests/test_mcp_parity.py records the exemption with that reason. A human checkpoint reachable from the MCP surface is not a human checkpoint — it is a second gate record with a longer name. gate skip is refused too: "the operator does not need to approve this" is not the agent's call.

What this is, precisely. An audit trail and a speed bump, not a security boundary. An agent with shell access can run ddflow approve itself, and no design here changes that — the tool does not control the machine.

The guarantee, as narrowly as it holds: no MCP tool records a human outcome, and a clearance carries the OS user and a human flag, so a forged one is visible in the log rather than indistinguishable from a real one. (gate.<id>.human is also refused by the config writer, because two MCP calls — flip the flag, then record — used to clear the gate with no shell involved. Declare human gates in .ddflow/gates.toml, which no tool writes.)

Opt-in: the shipped pipeline has no human gate, and a test keeps it that way.

Parallelism and coordination

$ ddflow next --phase P1
Ready (3 ready, 0 running, 1 blocked):
  P1.T1  persistent store
      writes: shortener/store.py, tests/test_store.py
  P1.T2  base62 encoder
      writes: shortener/encode.py, tests/test_encode.py

These are independent — run them in parallel worktrees.
  (blocked) P1.T3: deps — P1.T1 is open; P1.T2 is open

ddflow claim <ID> leases the item and binds it to a worktree.

What a claim holds is what you said it holds. claim --globs is recorded on the item, so no later heartbeat — live, or reviving a lapsed lease — resets the lease to the globs the item had before; the claim prints the globs it recorded. Globs are comma-separated or a JSON array, and the flag may repeat (every value is kept; over MCP a JSON array is read whole). A glob mangled by quoting is refused with the form every surface reads. update --globs replaces the item's globs, and a claimed item's lease with them; the result names what it dropped, and show prints the lease's globs beside the item's. --resources works the same way. ddflow wait --item X --globs .. judges the claim you are about to make rather than the stored globs, so READY means that claim will not be refused.

Files every item touches — a changelog, a research log, a regenerated config — would make every pair of items collide. Declare them, and many leases may hold them at once:

[lease]
append_only_globs = ["docs/CHANGELOG.md"]   # every item adds lines
shared_globs = ["configs/default.toml"]     # generated: regenerate after merging

A path inside either is exempt from the overlap check (claim, update, next, wait) and counts as covered at commit time for anyone holding a live lease. For each append-only glob in the committed config ddflow writes <glob> merge=union to .gitattributes — when the knob is set through ddflow config --set / ddflow_configure, and again on ddflow init / adopt after a hand edit — and prints the line; commit it with the config. A generated file never gets union (it would be interleaved): doctor notes a shared glob git has no merge driver for (a document items edit by hand in sections, like README.md, is merged right by git's default text merge: declare it with a README.md merge=text line in .gitattributes and the note stops), and reports an append-only glob git does not union-merge. Git decides which driver a file gets (check-attr), globs match as .gitattributes does, and a union line never overrides a narrower driver the project set for a file inside the glob: it is written before that line, so the narrower one keeps winning. Only the committed config writes union lines; a .ddflow/local edit syncs nothing.

A contended file is served first come, first served. An agent refused for an overlap used to have no place in line: it polled, and whoever polled first after the release took the file, so a hot file (the one every change touches) starved its longest waiter for hours. Now a live ddflow wait --item X, or a claim that was refused and is asked AGAIN, is a place in line (kept in .ddflow/local/waits/, this machine's, never committed). While the oldest waiter could claim its item right now, a younger or unqueued claim of an overlapping file is refused with exit 3: TC is reserved for agent-b (waiting since 14:02:11Z for TB, which needs the same files); their place is held for 300s after they could claim. next lists such an item as blocked instead of offering it, and wait --item stays blocked on it. The waiter's own claim passes, and it spends the place. This is a best-effort queue, not a lock: the registry is checked when a claim is asked, not inside the lease transaction, so two claims landing in the very same instant at a release can still race.

The rule cannot hold anyone up for long: a waiter whose process died or whose wait deadline passed is dead, and blocks nobody; a waiter that was woken and does not come back loses its place [lease].waiter_reservation_s seconds later (default 300; 0 turns the queue off); a waiter still behind another holder, or waiting for files disjoint from yours, reserves nothing against you; and the order is strict (older first), so two waiters cannot reserve against each other. A refused claim keeps its place as long as it is asked again within the window; a SINGLE refusal reserves nothing (an agent that takes the alternative the refusal offers never comes back, and must not hide the item from next or block its owner). The holder is told who it holds up: heartbeat and release name the waiters, and brief adds a "Waiting on you" section. Releasing globs while only gates are pending is deliberately not offered: until the branch is merged another agent editing the same files would only meet the conflict at merge. The honest remedy is a short claim: claim when you are ready to edit, run the gates promptly, and do not sit on a lease waiting for a slow review (record partial and merge instead).

When two clones disagree. A merge can bring in a rival definition of one id (two clones added it) or two claims on one item whose windows overlapped. ddflow records the contest instead of letting the later event silently win: show prints a CONTESTED block with every rival whole, next withholds the item, and doctor names it. A person settles it with one event:

$ ddflow resolve P1.T1 --keep alpha            # a claim: by its holder, or its event id
$ ddflow resolve P1.T1 --keep 3fa9c2 --refile-as P1.T1b   # a definition; the loser is re-added

Keeping a claim releases every claim that overlapped it and starts the kept claim's lease window now: a claim that had already lapsed is not stretched back over the gap, so a claim another agent legitimately made in the meantime is not retroactively contested. A holder's own lapsed claim, every displaced claim (no cap), and a contestant's late renewal are all kept in the record. --keep may also name the current holder when it met no part of the contest. An item that is not contested is refused.

A wrong worktree binding can be corrected. ddflow update <id> --worktree PATH rebinds the item, and your live lease on it, to that linked worktree and the branch checked out there. The field edit lands first and the rebind under the same lock, from a fresh read; an empty path is refused, and complete's tree check then follows the new tree.

If you are already in one, it adopts that one. Agent harnesses — Claude Code, Cursor — often isolate the agent themselves. Claiming from inside a linked worktree binds the item to that tree and branch rather than building a rival and telling you to leave the one holding your uncommitted work:

$ ddflow claim T1            # run from inside the harness's own worktree
claimed T1 (lease 1800s, renew every 300s)
  worktree: /work/agent-tree  (adopted — you were already in it)
  branch:   agent-work
  Carry on where you are.

ddflow never needed to have created the tree — it needs to know which tree an item is worked in, so recover can find stranded work and merge knows what to merge. An adopted tree is recorded as adopted, not created, so remove_on_merge will never delete something ddflow did not make. A tree already bound to another open item is refused: two items in one tree cannot be merged or recovered separately — unless that item has let go of it: merged, lease released, and nothing left in the tree (no uncommitted change, a HEAD its target already contains). worktree.adopt_existing = false restores the old behaviour; --no-worktree skips binding entirely. [worktree] local_files copies git-ignored, machine-local files (a .roborev.toml) from the primary into every tree a claim binds — one it creates, the harness tree it adopts, or the item's own tree on a re-claim — and never overwrites a file already there.

An item claimed --no-worktree has no branch of its own, so merge and review take one: the branch checked out in the worktree you run them from, or --branch <branch> from anywhere, and gate run runs in the worktree you stand in. Two places are not the item's and are refused rather than guessed: another open item's worktree, and — with worktrees on — the primary, whose working tree is nobody's in particular (other agents' event logs, anyone's uncommitted edit); there merge refuses, and review and gate run record UNAVAILABLE and say what to pass. A lone agent working in the primary says so with [worktree].enabled = false. The borrowed tree is never removed; uncommitted work in it is refused as for any merge; and paths the merge lands outside the item's globs are listed (outside_globs), since a borrowed branch can carry another item's commits too.

$ ddflow merge B-fix --branch agent-work
  no worktree of its own: 'agent-work' was landed, no tree touched.
merged B-fix (1c2d3e4f) into main

What merge reports and what it leaves alone. sha is the commit the base points at after the landing — the merge commit, or the branch head on a fast-forward; the forge's merge commit in PR mode — and branch_head is the merged branch's own head, which under a squash is not on the base at all. A tree you are standing in is kept (its directory is your shell's cwd; deleting it fails your next command), and ddflow cleanup --apply removes it once the item is complete. A merge that fails is aborted, so the primary checkout is never left mid-merge (doctor flags one that was left, with the git merge --abort that clears it). A branch with nothing ahead of its target is refused — it would record the item merged with nothing landed — unless --allow-empty.

A second agent is refused, and told what to take instead:

$ ddflow claim P1.T4 --agent gamma
P1.T4 writes 'shortener/store*.py' which overlaps 'shortener/store.py' held by alpha on P1.T1

You could take instead: P1.T2, P1.T5

Exit codes are the contract, and agents branch on them:

"Nothing is ready" and "everything is fine" are different facts. An agent that cannot tell them apart invents work.

The critical path is reported, because it, not the task count, sets the wall-clock floor — adding a fifth agent to a phase whose runtime is a four-deep chain buys nothing. It walks nested sub-tasks: an umbrella's open sub-tasks count as steps before it, and a phase another depends on contributes its chain. Items held back only by a cap (schedule.max_parallel_tasks, or a resource's capacity) are counted by status and brief, and the cap's message says when a slot is free. What next offers is conflict-aware: the free slots are filled in priority order, but an item whose globs overlap one already offered in the same answer is held back as a conflict naming that item ("globs overlap X ... offered in this plan"), and the next independent item takes its slot; status counts these apart from the cap-held ones ("N overlap an offered item" vs "N held by the parallelism cap"). ddflow wait sleeps until something is ready, but when every blocker needs a person — a dependency cycle, an expired lease under reclaim_policy = "report" — it refuses at once rather than sleeping to its timeout.

Log-derived flow signals

Besides the host, the adaptive parallelism controller takes what the project's own event log shows as input signals (the module computes them; the controller's sampling path that consumes them is a separate piece of the adaptive-flow work). ddflow/core/flowsignals.py computes each as a pure function of the log and an injected clock, against baselines taken from the same project's history, so no number is specific to a machine or a repository. Each ends at "now"; a signal with too little history is unavailable (None), which is neutral: it never lowers the limit and never justifies raising it.

A fresh project simply reports the first three as unavailable until it has history; nothing needs configuring. A merge git tried and failed (a conflict) is recorded as a failed merge gate outcome, which is what merge_failure_rate counts; a merge refused before git tried (target checked out elsewhere, another merge in progress) is not. Reviewer latency needs the review's start: gate.started is written for command gates today, so until ddflow review writes it too the ratio stays unavailable on a real log (filed as a follow-up).

Gitflow, pull requests and version tags

Two independent axes in [flow], because teams combine them freely:

Which workflow that makes, and what is not covered:

Each of these is a workflow choice: the operator sets it, or an agent records it, and when nobody does the default is applied at first use and followed from then on.

The agent's loop does not change. next → claim → work → gates → merge. What merge means changes with the repository's policy:

• merge in PR mode pushes the branch and opens (or updates) a pull/merge request, then releases the lease and parks the item in REVIEW. The agent is free at once and takes the next task — nobody waits for a human. It refuses to open a request for work whose own pipeline is unfinished: a reviewer's time is the scarce resource, and a refusal after the merge could no longer stop anything.

• pr sync turns what reviewers did back into queue state. next runs it for you while anything is in review (sync_on_next):

  the forge says	ddflow does
  merged	passes the merge gate on the forge's evidence (URL, merge sha), completes the item, retargets anything stacked on it, removes its tree
  changes requested	returns the item to the queue with the review text (bodies and line comments); brief leads with it; a re-claim resumes the same tree and a re-merge updates the same request
  closed	parks it for a person — a "no" is not something to retry
  approved, checks green	merges it (pr_merge = "on_approval", pinned to the approved head)

• Stacking keeps work moving through review. While T1 waits in review, a task that needs it may start on top of T1's branch (stack = true); its request targets T1's branch and is retargeted to the real base when T1 merges. ddflow never merges a stacked request first — that would land it unreviewed inside T1's merge. A task depending on two unmerged branches waits: one branch cannot sit on two.

• Who presses merge is pr_merge: on_approval (default — a person's approval is still required, and branch protection still applies), auto (ask the forge to auto-merge when its own rules are met), or human.

Gitflow. Tasks fork from develop as feature/ or bugfix/ branches (by tag); a task tagged hotfix forks from production and lands on production and develop. Merges never switch a checkout: a target that is not checked out is merged in a throwaway worktree, and one checked out in someone else's tree is refused.

Versions. version show reads the highest v1.2.3 tag reachable from the release branch and computes the next version from Conventional Commits (feat → minor, fix → patch, !/BREAKING CHANGE → major; below 1.0.0 a breaking change bumps minor) and from the tags of items finished since (breaking, feature, bug, hotfix). version cut tags it — annotated, with generated release notes. Under gitflow it cuts release/X from develop, merges it into production, tags it and merges the tag back into develop; in PR mode it opens the release request instead, and pr sync tags the merge commit once a person merges it and opens the back-merge request.

Changelog with the cut. version cut --changelog also writes the new version's section into CHANGELOG.md (the path of [export.changelog], default CHANGELOG.md) through ddflow export changelog: the Unreleased entries become ## [x.y.z] - date with the right compare link, and an empty Unreleased remains. The file is committed on the branch the tag names, so the tag's commit holds it; in gitflow with pull requests it is a commit on the release branch, part of the release request, and the request body is the same section (the tag, cut later by pr sync, carries its usual message); otherwise the tag message is the section. Mode whole (the default) rewrites the file; mode region rewrites only the marked region and puts the new section below it, updating the [Unreleased]: link line when the file keeps one. Nothing is written without the flag (the tag message is then the commit-derived notes as before), --dry-run shows what would be written, a hand-edited file or uncommitted changes to it are refused (exit 3) unless --force, and a git or log failure is exit 2. Trunk or maintenance cuts with pull requests refuse --changelog (no request to carry the commit). The MCP tool takes changelog and force. Bumping version files is a separate task.

An item counts as shipped on a line when its landing merge commit or the branch it merged (the merge's second parent) is reachable from it: a hotfix landed on production reaches develop by the back-merge as that branch, with production's merge commit nowhere in it. A squash or fast-forward landing is the one commit it made. A squash landing has no merge parent to follow, so it is also found on another line by its change: a commit there with the same patch-id (a cherry-pick, or the back-merge squash), or the back-merge <id> into ... commit ddflow wrote for it -- so a hotfix squashed onto production counts as shipped on develop once it is back-merged. Only the newest 1000 commits of the line are searched.

ddflow holds no token: it drives gh or glab, logged in as the operator, so every permission question is answered by the forge. A forge that cannot be reached is exit 2 — "could not ask" is never reported as "nothing changed".

[flow]
model = "gitflow"          # or "trunk"
integration = "pr"         # or "merge"
pr_merge = "on_approval"   # or "auto" | "human"
pr_reviewers = ["alice"]

The research behind this is RESEARCH R16.

Several release lines: fixes to older majors

Projects that keep older majors alive — main is 3.x while 2.x and 1.x still get fixes — declare them as release lines, oldest first. The newest line is always the current one and follows model as above; a maintenance line lands straight on its branch.

[flow]
current_line = "3"
port_strategy = "cherry-pick"      # or "forward-merge" (the default)

[flow.lines]                        # oldest first
"1" = "maint/1.x"
"2" = "maint/2.x"

• An item belongs to a line: task add T --line 2, or phase add P --line 2 and every task in it inherits it. A line that does not exist is refused, never read as "the current one".

• A fix for several lines is one command: task add FIX --lines 1,2,3. ddflow writes it where the strategy says and generates a port task FIX@<line> for each other line — an ordinary task with its own branch, gates and merge or pull request. A port starts only once what it carries has landed (review is not enough):

  port_strategy	the fix is written on	each port…	ports run
  forward-merge (default)	the oldest line	merges the previous line's branch into its own (and passes through every line in between — a merge cannot skip one)	one after another
  cherry-pick	the newest line	applies exactly what the fix landed (the target's before→after range, whatever the merge strategy) with a three-way apply	in parallel

• A conflicting port is work, not a failure. The claim leaves the conflict markers in the port's tree and names the files; the agent resolves, commits and carries on.

• Lines never collide. The same file on 2.x and on 3.x is two branches, so two agents may hold them at once; on the same line the glob check applies as always.

• Versions per line. version show --line 2 reads the highest tag reachable from maint/2.x; version cut --line 2 tags it there, and refuses a bump that would leave the 2.x major — a breaking change belongs on the current line.

Environment branches: promoting downstream

GitLab flow's environment branches — each mirroring what is deployed there — are declared in order, downstream of the current line's target (the base branch; under gitflow, production, so an environment receives what was released):

[flow]
environments = ["pre-production", "production"]
auto_promote = ["pre-production"]   # optional: continuous delivery to staging

• Work reaches an environment only by promotion, one step at a time. ddflow promote add pre-production files a task that merges main into pre-production; promote add production merges pre-production into production. So production only ever receives what the environment before it already has ("upstream first"). Nothing else targets an environment branch.
• A promotion is an ordinary task: claim it and its tree is made from the environment branch with the upstream branch already merged in (a conflict is left for the agent, like a port's). It runs gates.promotion_pipeline — unit_tests and merge by default, since what it carries already passed its own pipeline; add a human gate there for a person's sign-off on each deploy. It needs no cross-family reviewer: it authors nothing.
• With integration = "pr" the promotion lands through a merge request into the environment branch — the approval is the deploy approval — and pr sync completes it.
• One open promotion per environment, and none when there is nothing to carry (exit 2).
• auto_promote lists environments ddflow next promotes to by itself when the branch upstream moves. Empty by default: a deploy is the operator's call.
• ddflow promote status shows each environment's head, how many commits it is behind the branch upstream of it, and any open promotion.

Workflow choices: asked, recorded, defaulted on the record

ddflow supports several ways of working and never picks one silently. Each decision — model, integration, pr_merge, on_changes_requested, stack, port_strategy — is a choice, and its value comes from, in order:

1. the operator's config (.ddflow/config.toml or env), which always wins;
2. a recorded choice — ddflow flow choose port_strategy cherry-pick --reason "2.x has diverged", by the operator or by an agent the operator left it to, attributed in the log;
3. the default — applied the first time the choice matters (the first claim, the first pull request, the first fix filed across lines) and recorded, so the project keeps following it even if a later ddflow ships a different default.

Until then, a relevant choice nobody made heads ddflow brief under Open workflow choices, so an agent asks at the start rather than discovering at the end that the project wanted something else. ddflow flow show lists every choice with its value, its options, and who decided — config, a named agent or person with their reason, or "DEFAULT (nobody chose)". A recorded choice that the config file overrides is shown as such, never silently ignored.

Research: RESEARCH R17.

Many agents, one server: identity, state and sharing

Several agents and subagents sharing one queue is the case this tool is for. Here is exactly how that works, because each of these has a wrong answer that looks right.

Is it stateless?

The queue is. The connection is not, in exactly one respect.

The append-only event log is the sole source of truth, and every read re-derives state from it — fold(read_all()), from scratch, on every call. Nothing is cached between requests, so there is no stale projection, no invalidation, and no divergence between two agents' views. Restart the server mid-task and nothing is lost: it never had anything the log did not.

The one piece of per-connection state is who you are (below). It is deliberately not in the log, because it is a property of the caller, not of the work.

Who is calling?

By default, identity is derived from the working tree. That is right for one agent per worktree, and silently wrong for several agents in one tree — they all resolve the same path to the same name, their events merge into one stream, brief answers with a sibling's task, and reviewer-independence compares an agent with itself and passes. Nothing errors. There is no signal that can tell them apart, so identity is declared:

Under a declared identity a call uses what that agent holds, on every surface: an MCP tool called with a foreign as_agent asks from the primary and ddflow_claim creates that agent's own tree instead of adopting the connection's harness worktree, and ddflow --agent X claim does not adopt a tree another identity is working in -- one it has worked in under its tree-derived name, or one it ADOPTED (worktree.adopted names the adopter, so a parent that always declares --agent still owns its tree once it has claimed from it; a parent that has never claimed from its tree is not yet detectable, and the first identity to adopt a tree owns it). A reviewer written through ddflow_configure is recorded under the per-call as_agent. ddflow_setup writes into the tree the server stands in, as adopt does; adopt run from a linked worktree writes the project's tracked files into that worktree (the event log and hooks stay shared).

A name you set yourself is never suffixed, so the same DDFLOW_AGENT in two clones is still one agent to ddflow; ddflow doctor notes a shard whose clock goes backwards, which is what that leaves behind after a merge.

Innermost wins. ddflow_identify is idempotent, persists for the connection, and refuses a name that could not be a log filename — it becomes one, and refusing at declaration time means the caller reads the reason rather than discovering it at the first write.

If more than one agent works one tree at once, declare identity. Everything that attributes work depends on it. A subagent must not call ddflow_identify on a shared connection — that renames its parent — and passes as_agent instead; without it, two subagents claiming the same file are one holder, and a holder's own leases never conflict with each other.

Can one server serve several projects?

No — one server process serves one repository, fixed at start from --repo, DDFLOW_REPO, or the working directory. No tool takes a repo argument, and a test asserts none ever does. Point a second agent at a second project by running a second server; they are cheap, and the isolation is the point.

One project shared by many agents is the supported case — and the one that needs no special setup beyond declaring identity:

• Writes never conflict, and they never block readers. Each agent appends to its own log shard, so there is no shared file to overwrite and no merge conflict to resolve. Writers do serialise briefly: one repo-wide lock is held across the clock allocation, the append and its fsync. Short, but not nothing — per-agent shards remove file contention, not lock contention.
• Reads take no lock at all, so a read-heavy agent cannot be starved by a write-heavy one, and a reader can never block a writer.
• File ownership is coordinated by globs. claim refuses an item whose writes overlap one already held, and names what to take instead — exit 3, not a failure.
• Lessons, decisions, research and bug history are shared by construction: they are events in the same log, so one agent's finding is immediately visible to every other.

Locking, contention and measured cost

Measured on this machine, single process, full read_all() + fold():

Linear, converging on ~8.7 µs/event; the higher figure at small sizes is fixed per-call overhead, not the fold. A project with 20,000 events pays ~175 ms for a state-reading call. Search and recall do not pay this — they run off a SQLite projection rebuilt only when the log's head moves.

tests/test_mcp_load.py runs 12 concurrent agents through the real MCP surface and asserts no deadlock, no lost append, no repeated Lamport value within an agent, and correct attribution for every event — not for a sample. Its thresholds are environment variables (DDFLOW_LOAD_AGENTS, DDFLOW_WRITE_LATENCY_BUDGET_S, DDFLOW_GROWTH_TOLERANCE, …) because a load test with a hardcoded budget either flakes on a shared runner or is too loose to fail.

The deadlock bound is a hard timeout: a wedged lock does not fail, it hangs, and an unbounded hang reads as a broken CI runner rather than as a bug.

Crash recovery

An agent is killed. Nothing is cleaned up, because in a real crash nothing runs.

$ ddflow recover
1 recoverable situation(s); 1 may contain work:

!! P1.T1  [expired_lease]  was: delta
     worktree /repo/../.ddflow-worktrees/P1.T1
     INSPECT FIRST — 1 uncommitted file(s), 1 unmerged commit(s).
     `git -C .../P1.T1 diff main` then salvage,
     then `ddflow release P1.T1 --note salvaged`.

Four behaviours, each chosen against a specific way this goes wrong:

• While the lease is live, nothing happens. A dead agent is indistinguishable from a slow one until the lease expires, and guessing is how two agents end up in one tree.
• Recovery measures the tree — uncommitted files, unmerged commits — rather than trusting the recorded state. "Is there work in here?" is the only question that decides the remedy.
• An expired lease is never stolen silently, and recover --apply expires only trees it measured as empty.
• Adoption, not duplication: an agent resuming a recovered item gets the existing worktree back, not a second one beside it.

Reconstruction from logs alone

$ ddflow replay --out ./recovery-kit
wrote:
  recovery-kit/RECONSTRUCTION.md
  recovery-kit/QUEUE.md
  recovery-kit/LESSONS.md

RECONSTRUCTION.md is written as instructions to a fresh agent, not as a report about the past: every operator prompt in order, every research verdict, every lesson, the queue's shape — and every approach already tried and rejected, with the measurement that killed it.

It states its own limit, in the document: it reproduces the decisions, not the bytes. Model outputs are not deterministic, so replaying prompts will not recreate the original source. What it recreates is every input that produced it, which no other artefact holds.

The demo destroys an entire repository and rebuilds from 3.9 KB of JSONL, then checks nine specific fragments are present — including the operator's stated reason for a constraint, and the probe output behind a rejected design.

Secrets are redacted on the way in, not on the way out — the log is committed, so a scrub at read time is a scrub that git show walks straight past.

Lessons that check themselves

A lesson can name the mistake in code, not just in prose:

$ ddflow lesson add --title "Never swallow a bare OSError" \
      --rule "Catch the specific error; a broad except turns a loud failure into a silent one" \
      --pattern "except OSError" --globs "*.py"
lesson La0c745cc recorded — inventory: 2 site(s) now

$ ddflow lesson verify          # later, after somebody adds a third
La0c745cc: 1 NEW site(s): c.py: except OSError:

Exit 1 names the file. That is the whole design, and it comes from a failure worth repeating: on the project ddflow was extracted from, a count-based clone ratchet sat red for ~350 commits. It was advisory so it never blocked, it reported a number so every reader learned to skip it, and 24 new clones arrived through that gap.

A count says "worse" and never "which".

A number cannot be acted on or reviewed. A list can: a new entry is a line somebody opens, and a disappeared entry is progress — reported, and never a failure, because the inventory may only shrink.

Three details that decide whether a ratchet survives contact with a real repository:

• A site is <path>: <matched text>, not path:line. Line numbers churn on every edit above a site, which would invent a matching pair of "new site" and "fixed site" findings out of an unrelated change — and a ratchet that cries wolf is one that gets switched off.
• An uncompilable pattern is refused, not stored. An empty inventory reads exactly like a clean repository, and would ratchet every real occurrence away the first time it ran.
• Exit 2 when no lesson declares a pattern. Not a pass. A corpus with zero ratchets should not be able to report "all clear".

Vendored and untracked files are never sites — matches in code nobody owns are findings nobody will act on. Most lessons stay prose, and a prose lesson produces no findings at all.

Checking that the checks are working

Three questions ddflow asks about itself, all derived from the log and all reported by ddflow doctor. They exist because an unmeasured mechanism is indistinguishable from a missing one.

Can the queue's work actually be picked up? Every other check counts the items that are present; this one asks whether any of them can be started. An open phase with no task under it is work ddflow next will never offer — a note by default (schedule.empty_phase, since a project that files phases before breaking them down lives there on purpose) — and a phase whose tasks are all finished while the phase stays open is always a problem, because that is a queue held open by an item nobody can act on.

Tasks cannot go missing here, and that is a property rather than an untested gap: two hypotheses about how one could were probed and both refuted, and a test now pins the invariant so a future filter cannot quietly reintroduce it.

Does a gate ever say yes? A gate that fails on everything is worse than no gate: it trains the next reader to skip it. A gate at or above gates.rate_max_fail once it has gates.rate_min_runs decisive runs is reported as flaky or as measuring a moving target — re-running it will not converge. A skipped gate is not a run, because counting skips as failures would make an unconfigured gate look like a broken one.

Did the periodic passes ever fire? The mechanism you did not measure is the one that is not running. Because a cadence here counts completions rather than wall-clock, this is exact rather than estimated: since is the completions elapsed since the pass last fired, which is the same quantity ddflow cadence uses to decide due-ness — deliberately, because two measures of "is this behind" that can disagree is a situation nobody can reason about. A pass more than cadence.max_missed scheduled runs behind is reported. Being merely due is not a finding (ddflow cadence already says that), and running early is not one either.

All three are notes, not problems: a defect in the machinery that checks the work must not block the work.

Reading the log, and why it is never compacted

The log only grows, so every state-reading call used to re-read and re-parse all of it. Measured at 20,000 events, that read costs 115 ms — and the breakdown is the whole design argument:

Parsing dominates, and an append-only file guarantees the bytes already parsed have not changed. So EventLog.read_all re-parses only the appended tail, and re-hashes the bytes it is re-using to prove they are still the same bytes:

A command like ddflow doctor — which reads four times — pays the full cost once instead of four times.

Two knobs, [log]:

The validity check is a content check, and that is the whole design. The consumed prefix is re-hashed on every read — 1.1 ms to read plus 4.3 ms to digest, against the 97 ms of parsing it avoids. The first version used st_ino instead, on the reasoning that "a git merge writes a temp file and renames, so the inode changes". That is false:

$ git checkout -q other && stat -c %i .ddflow/events/a1.jsonl
218500670
$ git checkout -q main  && stat -c %i .ddflow/events/a1.jsonl
218500670

Git rewrites tracked files in place. So switching between two branches that had diverged left a warm server serving events from the branch you left, silently losing the ones actually on disk, with the tail read starting mid-line — and because Store.rebuild takes its fingerprint from the real file while taking its events from the cache, that wrong state was written into the SQLite index stamped as current, which a fresh process would not rebuild away. A content digest makes a rewrite, a truncation, a git checkout, a git merge, a delete-and-recreate and a torn tail all one case, so there is no list of mechanisms to keep current.

A torn final line from an append that died mid-write is reported by ddflow doctor and re-read until the writer completes it, never marked consumed. Every guarantee here is mutation-verified in tests/test_log_read_cache.py — including that the digest covers the whole prefix rather than a trailing window of it, which a smaller fixture cannot tell apart.

A log from a newer ddflow, and the link event kinds

Checkouts of one repository can run different ddflow versions, and a shard merged from a newer one may hold event kinds this code has no handler for. They are skipped, as an unknown config key is, but not silently: ddflow doctor adds a note naming each unknown kind and how many events were skipped (this log has events from a newer ddflow ... skipped: record.extended x2), and ddflow status carries them as skipped_kinds. Every number the older ddflow shows is computed without those events, so the remedy is the same as for a config key: merge main, or run the newer ddflow.

Two such kinds describe how records relate (decision D-no-duplicates). Add events (task.added, phase.added, bug.found, lesson.recorded, research.recorded, decision.recorded, memory.recorded) may carry extends, duplicate_of, related and dedupe (the recorded answer, its score and the candidates shown). A later record.extended is a verbatim addition to an existing record (text, who, when, score), and a later link.recorded is a link or a distinct dismissal. Both accumulate, keyed by event: two additions made at once by two clones both survive in any fold order, and an addition never replaces the record's own text (a bug's summary stays as written). ddflow replay renders both. The add paths write record.extended and the related back-link (see "The check every add runs"); no command writes a bare link.recorded yet.

The compaction that was declined

An event kind log.compacted was reserved for a retention pass that shrank the log. It has been removed, because the recipe it was reserved for cannot be implemented without breaking two shipped commands. Three probes:

1. A compaction survives a merge=union merge. One branch compacts, the other appends; the deletions stick. So union is not the obstacle.
2. Two divergent compactions merge to neither side's result, and out of Lamport order — the case union cannot resolve.
3. The decisive one. ddflow progress and ddflow loops read raw events, not folded state: progress.work pairs each lease.acquired with the next release across the whole history. Leases and gate outcomes are not PROVENANCE_KINDS, so keeping "the last state-bearing event per subject" leaves a lease.released with no acquire to pair with. A queue whose loop detector fires repeat_claims before compaction reports nothing after it, and six attempts become zero — and under [loops] on_detect = "block" that is a behaviour change, not just a lost report.

Growth is addressed by making the read cheap rather than the log short, which keeps it append-only and auditable. Beyond ~100k events the right answer is an on-disk state snapshot, not a shorter history.

Lessons, research and bugs

ddflow lesson add --title "Truncating a slug can leave a trailing separator" \
                   --rule "Strip separators AFTER slicing to length, not before."
ddflow lesson search "cutting a url short leaves a dangling hyphen"

Retrieval is BM25 over FTS5 and finds that entry despite no shared keyword. Probed against embeddings and found sufficient at lesson-corpus scale (R4); lessons.search_backend exists for when that stops being true.

Research entries must carry a verdict, and CONFIRMED/REFUTED are refused without a probe:

$ ddflow research --question "is it fast?" --verdict CONFIRMED
CONFIRMED requires a --probe (and ideally --probe-output): a verdict with no probe behind
it is an opinion. Use THEORETICAL and say why no probe was possible.

And a bug cannot be closed without the test that would catch it again:

$ ddflow bug fixed B1
a bug may not be closed without --regression-test naming the test that would catch it
again. Write the test, watch it FAIL against the unfixed code, then close.

That refusal is the whole mechanism by which the same bug does not ship twice.

A bug may also carry a title, a severity and a scope, the three things an upstream report or an exported BUGS.md needs:

ddflow bug found --summary "claim drops the lease on crash" \
                 --title "Lease dropped on crash" --severity high --scope ddflow

--severity is low, medium, high or critical (optional); --scope is project (the default) or ddflow, for a bug in ddflow itself rather than in your code. The same three are the title, severity and scope parameters of ddflow_bug_found. All are optional event fields: an old log folds unchanged and an older ddflow ignores them. The title takes part in the duplicate check, and show <bug> and ddflow similar display the title, the severity and a scope other than project. A later bug.reported_upstream event (url, number, delivery, sent_at, digest) records where a ddflow-scoped bug's report went; an older ddflow reports it as a skipped kind. Once a command to prepare that report exists, filing a ddflow-scoped bug ends with a one-line offer to run it.

Listing tasks, phases, bugs and research

ddflow task list, phase list, bug list and research list print one line per record -- id, state, title, owner and the date of the last change -- newest change first. They read the folded log and write nothing; ddflow show <id> is the whole record.

ddflow task list --phase P2 --state open        # what is left in a phase
ddflow phase list                               # id, state, done/total tasks, title
ddflow bug list                                 # OPEN bugs only
ddflow bug list --all                           # also fixed and invalid ones
ddflow bug list --state invalid --since 2026-10-01
ddflow research list --state refuted            # the state is the verdict, lower-cased
ddflow task list --tag web --limit 20 --json    # {rows, total, shown, truncated, ...}

Filters are --state, --phase, --tag, --owner (the leaseholder; not --agent, which is always who you are), --since (an ISO date or timestamp, compared with the last change) and --limit (default 50, at most 1000; a cut list says how many matched). A kind offers only the filters it has data for: bugs have no tag or owner, so bug list --tag x is an unknown option rather than a silent "everything" (and the list flags are refused on research when recording a finding). Titles and tags are redacted as the export documents are. Exit codes: 0 rows, 2 none matched (the message, and reason in --json, name the filters), 3 a refused value such as an unknown --phase. research list is the optional-verb form of research, like research add. These commands have no MCP tool yet: the one consolidated read tool for them is the next task of the viewers phase.

Searching everything

ddflow search "<text>" looks across tasks, phases, bugs, research, decisions, lessons, sessions (notes and end summaries), prompts (what the operator said) and the log (the payload text of the events no record source already holds). Read-only; it writes nothing.

ddflow search "worktree cleanup"                 # ranked: the same TF-IDF engine `similar` uses
ddflow search "B-mcp" --exact                    # case-insensitive substring
ddflow search 'lease\.(expired|released)' --regex --kind log --since 2026-10-01
ddflow search retry --kind bug,research --state open --phase P2 --owner alice --limit 10 --json

One line per hit: source, id, state, date, and a snippet around the match. A prompt, note or summary hit carries its session id. Filters are --kind (comma list of the sources above), --state, --phase (tasks, bugs and research under it), --owner (the leaseholder, session agent or event agent; not --agent, which is who you are), --since and --limit (default 20, at most 200). Snippets are cut from text already redacted like the exports, so a secret in a prompt is never printed and cannot be searched for. --regex refuses patterns that can take exponential time (a variable-length repeat inside a repeat (unless the two together run at most 8 times), an alternation inside a repeat that can run over 8 times, back-references, more than two unbounded repeats, over 200 characters) and looks at the first 2000 characters of each record under a 5 second budget. Exit codes: 0 hits, 2 none ("No matches ..."), 3 a refused request. No MCP tool yet: it joins the consolidated read tool of the viewers phase; agents use ddflow_recall meanwhile.

Cadences

Periodic whole-repo passes a per-task gate structurally cannot do. Due-ness is derived from completed work, so there is no state file to drift:

$ ddflow cadence
DUE: integration_tests — 5 tasks since last (every 5)
DUE: mutation_tests — 3 phases since last (every 3)

Record one with: ddflow cadence --ran <name>

Configurable: integration tests, architecture review, mutation testing, duplication sweep, lessons compression.

A phase will not close over an overdue pass. ddflow complete <phase> lists every due pass counted in phases (architecture review, mutation tests, lessons) as an unmet condition and refuses (exit 3). Run the pass and record it with ddflow cadence --ran <name>, or skip it on the record with ddflow cadence --ran <name> --note "skipped: <reason>". Task-counted passes stay advisory, and completing a task never asks. --force overrides (recorded), and every completion path, including a merged pull request, applies the check. The driver's phase close runs ddflow cadence first.

Exporting documents

The log already knows the roadmap, the open bugs, the decisions in force and what was done this week. ddflow export renders that as ordinary markdown, so someone who never runs ddflow can read it in a pull request. ddflow owns the format (a Jinja2 template per kind, rendered in a sandbox); you ask for a document, you do not write it.

What can be exported: roadmap (Now / Next / Later with progress), bugs (open, fixed, invalid), status, worklog (coalesced per item, grouped by day), sessions, decisions (an index), rules (decisions in force by governing path, lessons, the enforced workflow) and changelog (Keep a Changelog form, from finished items, fixed bugs and version tags). Replay is not a kind (ddflow export replay is refused): it carries paths and addresses. Output is deterministic: the same log gives the same bytes, with no clock in the body.

Every default target is at the repo root; [export.<doc>].path moves it. rules exports decisions in force, lessons and workflow settings: imported rulebooks are not exported. The changelog takes each finished item (and each fixed bug that carries a changelog line) and files it under Added / Changed / Deprecated / Removed / Fixed / Security by the first rule that answers: the explicit --changelog "Added: ..." line given to ddflow complete (see Changelog line), then the Conventional Commit type of the item's landing commit, then the item's tags (breaking, feature, bug, hotfix), then a (fixes bug X) title suffix, else Changed. --changelog skip keeps an item out. An entry lands under the OLDEST version tag whose history contains it, otherwise under Unreleased; one with no merge commit is placed by date and says so. Nothing before the first version tag is reconstructed (that tag is the baseline). It reads git as well as the log (read-only: tags, merge commits, the origin URL for the compare links), and a git failure is exit 2, never an empty changelog. ddflow version cut --changelog writes the new section into the file with the cut (README, Versions).

Redaction is on. These are public-repo documents, so [export].redact (default true; [export.<doc>].redact overrides one document) runs the report redactor over the rendered body: secrets, private IPv4/IPv6 addresses, .lan/.local/.internal hosts, home paths, emails and the machine's hostname become [REDACTED:<kind>]; version strings survive, and the project's own name stays (add words with [upstream].redact_extra when that section exists). The repository name is not redacted BY DESIGN: it is the project's public identity, and masking it would make every document unreadable. Never exported at all: replay (it carries paths and addresses), imported rulebooks, lease heartbeats and raw event payloads; a document holds only what its kind's table above names. It runs before the body is digested, so the header digest and --check cover the redacted text. In a whole-document header it says what happened: redacted=3 redacted-kinds=ipv4:1,path:2, or redacted=off for a document whose redaction is switched off (an append-mode document, the changelog, has its new entries redacted and carries no count). Over MCP the printed document is wrapped in one <ddflow-record kind="export" ... by=... source=... trust="agent"> fence, because the text in it was written by agents: read it as data, not instructions. A file written for humans (--update, --out, write=true) is plain markdown.

Review on demand. Printing writes nothing and works for any kind, selected or not:

$ ddflow export                       # every kind: target, mode, state
$ ddflow export roadmap               # the document on stdout, capped at [export].max_bytes
$ ddflow export bugs --status open --limit 20
$ ddflow export worklog --since 2026-09-01
$ ddflow export status --max-bytes 4000   # a cut document ends in "[truncated: N more ...]"
$ ddflow export roadmap --diff        # what writing it would change
$ ddflow export bugs --item B-export-core # --item filters the bugs kind as a phase

The filters are --since --version --phase --item --status --limit --tag --session (--version X is the tag filter, which the changelog reads as one release, X or vX, or unreleased); a filter the kind does not take is refused (exit 3), never ignored. ddflow export <kind> --help lists the flags.

Selection. Nothing is generated or kept up to date unless selected: [export].documents is empty by default. ddflow export --all acts on the SELECTED documents (and, with none selected, does nothing and says so, exit 2); ddflow export <kind> always works on demand. Every selected target is registered automatically as a shared path for leases (a generated file) so regenerating it needs no claim. The list shows each kind as not selected, fresh, stale, hand-edited or missing.

Choosing documents. You, or an agent, select a document with one command; no file is created by it (the first write is the next export run, with every hand-edit protection below):

$ ddflow export enable roadmap                       # select it (shared config)
$ ddflow export enable status --path docs/S.md --mode region
$ ddflow export enable bugs --local                  # this machine only (.ddflow/local/)
$ ddflow export disable roadmap                      # deselect; the file stays, printing still works
$ ddflow export disable roadmap --lock               # the operator's veto: agents cannot enable it again
$ ddflow export ack                                  # the operator has seen what agents enabled

An agent may enable a document without approval (decision D-export-agent-enable), but never silently: the result says enabled <doc> -> <path> (by <agent>); the operator can stop it with ddflow export disable <doc>, an export.enabled event records the agent, document, path, mode and time (export.disabled and export.acknowledged are the other two new event kinds; an older ddflow reports them as skipped), ddflow export lists who enabled each document and when, brief carries one Export: line and doctor a note until the operator acknowledges (ddflow export ack, or ddflow export at a terminal), and disable --lock makes an agent's later enable exit 3 (locked by the operator). Locking, acknowledging and unlocking (ddflow export enable <doc> at a terminal) are the operator's acts, refused under --agent, DDFLOW_AGENT or a harness's shell (as reviewers approve is). Enable edits the selection through the same guarded write as config --set, and an append target is registered as an append-only path.

Templates are yours to change. The format of every kind is a Jinja2 template; resolution is [export.<doc>].template, then .ddflow/templates/export/<kind>.md.j2, then the shipped default.

$ ddflow export eject roadmap        # copy the shipped template into .ddflow/templates/export/
$ ddflow export eject roadmap --force  # replace an edited copy with the shipped default
$ ddflow export validate [roadmap]   # render the selected documents; errors with file and line

eject is idempotent and never overwrites an edited copy without --force; the copy starts with a comment ({# ddflow-shipped: <digest> -#}, renders to nothing) recording the shipped text it came from, so a ddflow upgrade never touches it but validate and doctor say when the shipped default has moved on, and an unedited older copy is refreshed by a plain eject. validate renders every selected document (or the one named) against the current data and exits 2 on any template error, naming the file and line; it writes nothing.

The template contract. A template sees ONLY the plain data of its kind (strings, numbers, lists, dicts), plus schema_version and a few filters; ddflow adds fields within a schema_version and bumps it only when it renames or removes one, so an edited template keeps rendering across an upgrade. The kinds' top-level variables:

(The authoritative list is the kind's data() in ddflow/services/export/kind_<name>.py; the shipped template in ddflow/templates/export/ shows each variable in use, and ddflow export eject <kind> gives you a copy to read.) The filters, besides Jinja's built-ins: md_escape (escape what would change rendering in free text -- emphasis markers, pipes, backticks, line-start list/heading/quote markers, tag-like < -- and leave identifiers such as run_nemo_run alone), wrap(width), date (the YYYY-MM-DD of a timestamp), truncate(limit) and bar(done, total, width) (a [###.......] bar). Templates run in Jinja2's sandbox (strict undefined variables, no autoescape): no attribute walks to __class__/__globals__, no open, no imports, a time limit and ceilings on output and on one operation's size, because a template may arrive with a cloned repository. Treat the ceilings as containment, not as a promise that a hostile template is harmless. A template error is exit 2 naming the file and line. The generated header and the body digest are added AROUND the template's output, so no template can remove hand-edit protection; and the redactor runs on what a template produced, so customising cannot leak what redaction would have caught.

Upgrade drift. eject records the shipped text it copied ({# ddflow-shipped: <digest> -#} on line 1). When a later ddflow changes the shipped default, your copy is left alone, and validate and doctor print a note: export template ... was ejected from an older shipped default. To pick up the new default: copy your file aside, run ddflow export eject <kind> --force (it replaces your copy with the shipped one), diff the two, and re-apply your edits.

Worked example: a changelog in your own words. Suppose release notes should name the item behind each line, drop the "placed by date" marker, and open with a product intro instead of the Keep a Changelog boilerplate.

$ ddflow export eject changelog
ejected changelog -> .ddflow/templates/export/changelog.md.j2
$ $EDITOR .ddflow/templates/export/changelog.md.j2

Two edits in that file. Replace the entry line

- {{ e.line }}{{ " _(placed by date, no merge commit)_" if e.by_date else "" }}

with

- {{ e.line }} ({{ e.id }})

and replace the paragraph that begins All notable changes to this project with Release notes for Acme Billing. Then check and review before writing anything:

$ ddflow export validate changelog      # renders it; errors name file and line
changelog: ok (.ddflow/templates/export/changelog.md.j2)
$ ddflow export changelog | head        # the new format, on stdout
$ ddflow export changelog --diff        # what --update would change in CHANGELOG.md
$ ddflow export changelog --update

Because the file's digest covers its body, the next ddflow export --all --check calls the old CHANGELOG.md stale until --update rewrites it with your format: that is the check working, not a fault. Commit .ddflow/templates/export/changelog.md.j2 with the project so every clone and CI renders the same bytes. To try a template without installing it, use ddflow export changelog --template my.md.j2 (writes nothing). Because a schema_version change is the only thing that renames a variable, a template written against version 1 keeps working, and a variable you misspell is an error (strict undefined), never an empty line.

The three update modes, per [export.<doc>].mode: whole (the file is generated; its header carries the kind, the ddflow version and a digest of the BODY, so hand edits are detected and --check regenerates and compares exactly), region (only the text between <!-- ddflow:begin doc=<kind> ... --> and <!-- ddflow:end doc=<kind> --> in a hand-written file; the rest is kept byte for byte) and append (a kind that can grow a log, today the changelog: entries are added after the last exported event). Writes are atomic under a lock. A hand-edited generated file, an unmarked file or an edited region is never overwritten without --force (exit 3, with the diff); paths outside the repo, symlinks and .git / .ddflow are refused.

$ ddflow export roadmap --update          # write the target; asks to confirm on a terminal
$ ddflow export roadmap --out docs/ROADMAP.md
$ ddflow export --all --check             # exit 1 if any selected document is stale
$ ddflow export --all --update --yes
$ ddflow export roadmap --template my.md.j2   # render once with another template; writes nothing

Enforcement. The pre-commit check behind [enforce].generated_views (block | warn | off) covers exported documents as well as the rendered views: a staged file whose FIRST line is a ddflow:generated doc=<kind> header of a known kind must equal a fresh export (body and kind compared, not the ddflow version), the event log must be fully staged beside it, and a stale or hand-edited document is named with ddflow export <kind> --update as the remedy. A hand-written file that only holds a marker region has no such first line, so it is not treated as a generated file; an append-mode log (last= in its header) grows by design and is skipped. Every SELECTED whole or append export target is also excluded from [enforce].stale_docs (shared_files.doc_exclude) and from docscheck (generated documents are judged by export --check; a region target's hand-written text is still scanned), every selected target is a shared path (no claim needed), and draws no "no merge strategy" note from doctor; doctor instead reports a selected target that is stale or hand-edited as a NOTE.

Exit codes: 0 done or fresh, 1 stale (--check), 2 could not run (unreadable log, template error, nothing selected) or declined at the prompt, 3 refused. A failure is never an empty clean document.

MCP. ddflow_export with no doc lists the kinds; with a doc it returns the markdown and says truncated: true (with truncated_more) when it was cut, bounded at 60,000 bytes whatever max_bytes asks. It writes only with write=true AND a repo-relative path (path alone is refused); diff and check compare against a path. It has no force, no update and no ad hoc template: an agent never overrides hand-edit protection or points the renderer at an arbitrary file. action = list, enable, disable (with doc, and path / mode for enable) or validate does the matching selection or template check; an enable over MCP is always an agent's (it names the agent and the stop command), and MCP cannot lock, acknowledge, eject or edit a template. It is in the all tool tier only.

The [export] knobs (5 of the 150): documents (the selection, default []), redact (default true), max_bytes (the stdout / MCP cap, default 60000; a written file is never capped), refresh (off | merge | phase_close | docs_gate, default off) and tables (the per-document tables below). Each document may have a table:

[export]
documents = ["roadmap", "status"]

[export.roadmap]
path = "docs/ROADMAP.md"      # default: ROADMAP.md at the repo root
mode = "whole"                # whole | region | append
template = "tools/roadmap.md.j2"
filters = { limit = 50 }

Unknown keys are skipped with a warning (a newer release's config does not stop an older checkout); ddflow config --set refuses them. ddflow help export has the same material.

Automatic refresh. refresh says when a SELECTED whole-file document regenerates itself; set it for all documents in [export] or for one in [export.<doc>] (the per-document value wins, so refresh = "off" there opts one document out):

Every mode keeps the same guards: a hand-edited or unmarked file is never touched (the result says skipped), a region or append document is not refreshed automatically, a refresh that fails (unreadable log, template error) is reported and never fails the merge or the completion, and with nothing selected for that trigger it does nothing and says so. The ddflow merge and ddflow complete output carries an export_refresh result (--json: documents, actions, digests) and prints a one-line summary. ddflow cadence lists export_refresh as due when an opted-in document is stale.

Keeping AGENTS.md true

ddflow adopt writes a managed block into AGENTS.md (and CLAUDE.md, and each agent's native rules file). That block is what tells an agent it must claim an item before editing — and every coordination guarantee here rests on that, because an agent that does not claim has its work destroyed by a parallel one.

Nothing used to check it again. Adoption is judged by .ddflow/config.toml existing, so a deleted AGENTS.md, a block someone stripped, or a block written by an older ddflow all left the agent reading rules that were absent or wrong while every surface reported the project as adopted. Adoption is a config file; the instructions are a separate fact.

Cursor does not really follow AGENTS.md, and it is not alone. Its precedence is Team Rules > Project Rules > User Rules > .cursorrules > AGENTS.md, so .cursor/rules/ddflow.mdc is what actually binds — which is why adopt writes it. That file is checked too, for every agent the project was adopted for (read from the driver deltas on disk, so a Claude-only project is never asked for a Cursor rule).

It carries the same block with binding frontmatter, and alwaysApply: true is part of what is verified: a rule with alwaysApply: false exists, reads perfectly, and may never be loaded — which for claim-before-you-edit is the same as not having it, and strictly worse than drifted text. It is reported at the severity of missing, not of stale.

Five states are detected — current, stale (drifted from what this version writes), no_block (file there, block gone), not_binding (native rule that will not apply), missing — and reported on three surfaces:

The two surfaces repair it differently, on purpose.

• From a shell, the operator is right there: ddflow adopt --refresh-docs rewrites the block (plain ddflow adopt does too, and also rewrites MCP launches). It replaces only what is between the DDFLOW:BEGIN/DDFLOW:END markers and leaves the rest of your file alone, and re-running it is a no-op. ddflow init reports the problem and does not write — writing prose into your AGENTS.md is not what init was asked to do.
• Over MCP, ddflow does not touch it. The handshake tells the agent to show the operator what is wrong and call ddflow_setup only if they agree. It is a file in their repository, usually with their own prose around the block, and rewriting it is not a decision a tool gets to make on their behalf — the same rule as companions ("propose; never install") and the human-approval gate.

Surviving a compaction

The instruction block reaches the model once, at connect. After a context compaction it may retain none of it, and MCP has no server-to-client primitive for injecting context — the three that exist (roots/list, sampling/createMessage, elicitation/create) all go the other way or ask a question. Three things already survive:

• the AGENTS.md / CLAUDE.md sections ddflow adopt (or ddflow_setup) writes, plus each agent's native rules file — the client re-reads its own rules, so this is the durable channel;
• the commit hook, which refuses a commit with no item trailer and says what to add. Enforcement at the moment of the act needs no context at all;
• ddflow help <topic>, which the agent can ask for — if it thinks to.

What none of those do is speak up unprompted. A footer on tool results is the only channel that is guaranteed to be heard again, because an agent driving ddflow calls tools continuously:

ddflow: left undone in this project —
  · 1 bug(s) still open: B1 — close with `ddflow_bug_fixed` (it requires the regression test) or say why not
  · 2 gate(s) skipped, not run: T4.critic, T4.standards — run them, or leave the skip on the record deliberately

It is not a banner, and the difference is the whole design. A fixed reminder appended to 63 tools is trained out inside a session and costs tokens on every call. This one:

• names what happened, never restates a rule — an id, a count, and the call that discharges it;
• stops once the thing is dealt with, so it cannot be trained out by repetition;
• says nothing at all when the project has nothing outstanding — not a cheerful "all clear", which is the same thing readers learn to skip;
• is cadenced: at most once every every_calls calls and every_seconds seconds, so a burst of calls is not a burst of footers;
• cannot break the call it rides on. It is a courtesy on top of an answer, appended after the body, and a failure inside it is swallowed. content[0] is still the structured result.

[reinstruct]
enabled      = true   # false silences it entirely
every_calls  = 12
every_seconds = 240
max_items    = 3

What it currently notices: bugs found and never closed, gates skipped and never revisited, and work finishing with no lesson ever recorded (after [lessons] reflect_after_items, so one task is not reported — the pattern is, and the threshold is a knob because where the line sits is a judgement).

Keeping session-start cost flat

ddflow brief --phase P2

Returns, inside session.brief_max_tokens (default 1200): recoverable work first, then the current item and its remaining gates, then what is ready, then why everything else is blocked, then the handful of past lessons ranked against this task's text.

This replaces reading the project's rule and lesson corpora. The budget is enforced by truncating from the bottom, so the safety-critical head survives a squeeze — and a project's opening cost stays roughly constant as its lesson corpus grows.

Agent portability

One canonical driver, templates/drivers/implement-phase.md, plus a delta per agent covering only what genuinely differs: how iteration continues, how to ask the operator, how to spawn a subagent, file-reference syntax.

Deltas rather than copies, for a measured reason: on the project this was extracted from, a reworded per-agent duplicate of the driver silently accumulated three instructions that were false at the time of writing while missing four gates the canonical file had gained. A delta removes the surface that can drift instead of policing it.

Both surfaces are one implementation — the MCP server maps each tool onto the same cli.main() call in-process, and two tests plus a demo step assert they cannot diverge.

Keeping the two surfaces honest

Every CLI command is reachable over MCP — that is the point of the tool list, and it is the requirement that an operator in a chat window, possibly driving a remote agent, can do everything a shell can. Three ratchets keep it true, and each one was added after the previous one turned out to be too shallow:

A fourth ratchet bounds the cost of that surface: the whole tool list is sent to the model on every session, so tests/test_mcp_tool_budget.py fails if the compact tools/list exceeds its byte budget (about 93 KB for 91 tools, down from 119 KB) or if the shared as_agent / relation / check_only descriptions are repeated at length on any tool. Their full text lives once, in ddflow_identify and the handshake instructions.

Tool tiers. A client that loads every tool schema up front still pays that ~93 KB, so [mcp].tools = "core" | "standard" | "all" (env DDFLOW_MCP_TOOLS; default all) chooses which tools tools/list advertises: core is 32 tools, 39 KB (the daily loop: brief, next, claim, heartbeat, gates, complete, merge, status, show, recall, bugs, lessons, decisions, sessions, identify, task add/update, wait, review, help, pr sync, similar, setup), standard is 63 tools, 67 KB (core plus the commonly used rest), all is every tool, byte-identical to before. It is a start-time choice and only about what is listed: a tool outside the tier is still callable by name, ddflow_help and the connection instructions say what the tier hides and how to widen it, and listChanged stays false, so change the knob and restart the server. An unrecognised value is refused on write and lists everything at start. The tier sets are one structure in ddflow/surfaces/mcp.py (CORE_TOOLS, STANDARD_EXTRA_TOOLS, FULL_ONLY_TOOLS) and tests/test_mcp_tool_tiers.py fails until every new tool is placed in exactly one. The parity ratchets read the whole registry, so they never depend on the tier.

The flag ratchet derives its own input from the parser rather than a hand-written list — its first version carried eleven tools and was blind to remove --force for exactly that reason. Omissions are allowed, but each must be an entry in FLAG_EXEMPTIONS with its reason, so "we chose not to expose this" and "nobody noticed" stop looking alike.

A fourth pins something subtler: whether a tool returns JSON or prose is a decision, not an accident. Some tools deliberately return prose — brief, gate status and replay exist to hand the model an instruction or a narrative, and JSON-encoding a paragraph so the client can decode it again helps nobody. But decision add returned JSON while task add returned prose for no reason either could state. Each prose tool now carries its justification in PROSE_TOOLS.

Command reference

ddflow adopt [--agents ...]     install into a project, for one or more agents
ddflow adopt --refresh-docs     rewrite only the driver docs + rules blocks (doctor says when)
ddflow init                     create .ddflow/ only

ddflow phase add <id> [...]     add a phase
ddflow task add <id> --phase .. add a task
ddflow update <id> [...]        change title/body/needs/globs/tags/priority (--globs REPLACES)
ddflow update <id> --worktree P rebind the item and its lease to the linked worktree at P
ddflow resolve <id> --keep X    settle a contested item: keep one definition or one claim

ddflow next [--phase P]         what may start now       (2 = nothing actionable)
ddflow claim <id> [--globs ..]  lease + create worktree  (3 = refused)
ddflow heartbeat <id>           renew a lease
ddflow release <id>             give it up

ddflow gate status <id>         pipeline position + the next gate's instruction
ddflow gate run <id> <gate>     execute a command gate, record its evidence
ddflow gate record <id> <gate>  record an agent gate    (--outcome, --reason, --model, --reviewer-model, --reviewed-sha)
ddflow gate skip <id> <gate>    skip, with a mandatory reason
ddflow approve <id> <gate>      a PERSON clears a human gate  (no MCP equivalent)
ddflow approve .. --reject      ...or refuses it, with --reason
ddflow gate verify <id> <gate>  prove the gate CAN fail  (1 = it cannot)
ddflow tests [--item <id>]      tests the change reaches + a parallel command  (2 = none)

ddflow merge <id>               merge from the primary checkout, no checkout
ddflow merge <id> --allow-empty land a branch with nothing ahead of its target (refused otherwise)
ddflow merge <id> --branch <b>  an item claimed --no-worktree: land <b> (default: your tree's branch)
                                ([flow].integration=pr: push + open/update a PR instead)
ddflow pr sync [--item]         what reviewers did: complete / reopen / park / merge (2 = forge unreachable)
ddflow pr status               every item's request, from the log (no forge call)
ddflow version show            current and next version, why, release notes (2 = nothing new)
ddflow version cut [--push]    tag it (gitflow: via release/X, or a release PR)
ddflow version cut --changelog  also write the version's CHANGELOG.md section (--force over a hand-edited file)
ddflow version show|cut --line L    the same, for a maintenance line (keeps its major)
ddflow task add <id> --lines 1,2,3  a fix for several release lines: ports generated
ddflow promote add <env>        file a promotion one step downstream (2 = nothing to carry)
ddflow promote status           each environment: head, behind upstream, open promotion
ddflow flow show                how this project works: model, lines, every choice + who made it
ddflow flow choose <knob> <v>   record a workflow choice, with --reason
ddflow complete <id>            finish        (3 = unmet conditions, all listed)
ddflow block <id> --reason ..   mark blocked (a DONE or ABANDONED item is refused, exit 3, unless --reopen)
ddflow abandon <id> --reason .. stop work on an item without completing it
ddflow remove <id> [--force]    take an item out of the queue (recorded, not erased)

ddflow brief [--item|--phase]   budgeted session-start pack
ddflow board / show <id>        human views (show also takes a bug id; --json board is JSON)
ddflow wait [--item|--phase]    sleep until something is ready  (--globs: judge the claim you will make)
ddflow render                   regenerate docs/ddflow/*.md

ddflow help [topic]             what this is, what it can do, the workflow
ddflow workflow                 the rules this project runs by  (1 = incoherent)
ddflow workflow pipeline ...    set the gates a task or phase passes
ddflow workflow gate ...        define or change one gate
ddflow workflow drop <id>       take a gate out of the pipelines
ddflow import [--apply]         propose an existing project's work  (2 = nothing)
ddflow import --verify          is the import still true, and did anyone finish it?
ddflow history [--item|--kind|--agent|--tail]  one timeline of everything that happened (2 = nothing)

ddflow lesson add|search        capture and retrieve lessons
ddflow research --verdict ..    record a finding (probe required for CONFIRMED/REFUTED)
ddflow bug found|fixed          regression test required to close (--regression-test repeats)
ddflow task|phase|bug|research list   one line per record, filters, --json (2 = none match)

ddflow session start|prompt|note|end     provenance logging (prompt/note: id optional)
ddflow session adopt-orphans             attach id-less prompts/notes to a session
ddflow session list|show <id>            read sessions back (newest first; --state --owner --since --limit)
ddflow search "<text>" [--exact|--regex]  search every record, session, prompt and the log (2 = no match)
ddflow replay [--out DIR] [--verify]     reconstruct from the log

ddflow recover [--apply]        find crashed agents' work   (2 = nothing)
ddflow cleanup [--apply]        classify ddflow worktrees/branches; --apply lands the safe ones
ddflow pins <file>              which text of an instruction file a test pins, before you compress it
ddflow precommit [--write]      propose a .pre-commit-config.yaml for this repo's stacks (writes only with --write)
ddflow doctor                   integrity + health
ddflow rebuild                  re-derive the index
ddflow export                   list the document kinds with target and state
ddflow export <doc> [filters]   print one (--since --phase --status --limit --max-bytes ...)
ddflow export <doc> --diff|--check|--update|--out P   compare or write (--force, --yes)
ddflow export --all --check     act on the selected documents ([export].documents)
ddflow cadence [--ran NAME]     which periodic passes are due  (2 = none)
ddflow config --explain         every knob, its value, its source and its docs
ddflow config --append-toml ..  add config without a shell editor (validated first; --local: not committed)
ddflow reviewers detect|list|test   find and check cross-family review endpoints
ddflow review <id> --gate ..    run the configured reviewer, record the evidence
ddflow review <id> --gate G --chunk N   re-review only chunk N of the recorded review
ddflow review triage <id> --gate G --finding N --refuted|--confirmed --probe ..
                                record what became of one finding
ddflow reviewers approve [name] a PERSON approves a tool-written reviewer  (no MCP tool)
ddflow mcp                      run the MCP stdio server

Every one of these is reachable over MCP, and a test enforces it. One tool goes the other way and has no CLI equivalent, because it has nothing to mean there:

ddflow_identify(agent=...)      declare who you are ON THIS CONNECTION (MCP only)

A CLI invocation is one process that exits, so it says who it is with --agent and the question does not outlive the command. An MCP connection is a session, so identity is declared once and persists — see Who is calling?.

Configuration

150 knobs across 21 sections, every one documented in place:

$ ddflow config --explain --filter lease
lease.ttl_s = 1800   [default]
    Seconds a lease stays valid without a heartbeat. After this it is EXPIRED and
    reclaimable. Longer = fewer false expiries when an agent is deep in a slow gate;
    shorter = faster recovery after a crash.

Resolution: dataclass defaults → .ddflow/config.toml → DDFLOW_<SECTION>_<KNOB> env. An unknown knob in a file is never a silent drop: ddflow skips it, so a config written by a newer ddflow does not stop an older checkout, but it warns on every command and ddflow doctor reports it as a problem (a typo, or a config newer than this code — merge main). ddflow config --set refuses an unknown knob or an invalid value outright, before writing. A map or list knob (list[str], dict[str, str]) given by environment as JSON refuses a non-string element instead of casting it (null is not the string "None"). A test asserts every knob carries documentation, so the reference cannot rot.

What is automated, and what is not

The honest split, because a tool that claims to automate judgement is lying about the part that matters.

Automated — happens without anyone remembering it:

• The handshake briefs the agent. On connect, the MCP server injects the live state: the pipeline every task must pass, work recoverable after a crash, what is ready, which companions are missing, and what to do about each. It is a template (ddflow prompts eject mcp_instructions), so the workflow is text you edit, not code you fork.
• Gates are enforced, not suggested. ddflow complete refuses on a required gate that has not passed, on open sub-tasks, on a silent gate under require_outcome, on a requirement no pipeline runs, and on a reviewer from the author's own family. Refusals list every unmet condition, not the first — an agent that cannot see how many more are coming reaches for --force.
• Conflicts are refused at claim time, by glob overlap, with an alternative named.
• Dependencies gate readiness. ddflow next withholds a task whose needs are open and says which.
• Bugs are offered before features. A task tagged bug/fix/hotfix (the [flow] bugfix and hotfix tags), or named by an open bug record, comes ahead of every feature in ddflow next and gets a free slot first, so a standing bug is fixed before more work is built on it. Priority orders each group; [schedule] bugs_first = false orders by priority alone.
• Gate evidence records which tree and how much — a working-tree fingerprint plus files/lines changed — so a pass names what it passed on. If the tree moves afterwards, complete warns that the evidence describes source nobody is shipping.
• Crash recovery: ddflow recover finds worktrees whose lease expired, so an interrupted agent's work is found rather than lost.
• Cadences (ddflow cadence) tell you which periodic passes are due — bug hunts, dedupe, lesson compression — from the log, counted in completed work; and, for a rule like "a bug hunt every week", by the calendar ([cadence] every_days = ["bug_hunt=7"]: never run means due now). The SessionStart hook lists what is due. A phase will not complete while a phase-counted pass is overdue.
• Post-merge review: ddflow review <item> --commit <sha> reviews one landed commit against its first parent — a merge as what it brought in — when the branch is gone.
• A commit hook (ddflow hooks install) can refuse an unclaimed edit outright, and refuses a staged ddflow render view that the log no longer regenerates byte-for-byte — hand-edited, or stale ([enforce] generated_views). It also reports a doc line still naming an identifier, file or default the commit removes or renames ([enforce] stale_docs, doc_globs, doc_exclude; warns by default). It refuses a commit on a branch whose base changed a rulebook (AGENTS.md, CLAUDE.md, each agent's native rules file, the driver docs) since it forked ([enforce] stale_rules), and warns past max_behind commits behind ([enforce] behind); the commit concluding git merge <base> is exempt, and the session hook only informs. Its commit-msg sibling requires an Item: trailer when [enforce] require_item_trailer is on — or the project's own keys (item_trailer_keys = ["Phase", "Phase-ships"]); merges are exempt. The value must be the id of an item in the queue (any state but removed): a mistyped id is refused with the nearest real ones, and a queue the hook cannot read is exit 2, never a pass. A key that marks a commit shipping no item takes only its declared words instead (trailer_waivers = { "Phase-ships" = ["none", "recon"] }). It also refuses any trailer named in [enforce] forbidden_trailers (e.g. a tool-attribution line), merges included, for every agent and every route that runs git hooks -- which a harness-side hook reading only the command text cannot promise. The hook's refusals say what to do. A lapsed lease on the item whose tree you commit from is named with when it lapsed and its remedy (ddflow --agent <holder> heartbeat <id> for its holder; for anyone else recover --item, release --note salvaged, claim). When nothing declared who is committing, it says the name was derived from the tree and how to declare one. A clean merge commit — the automatic one concluding git merge <base>, or a squash that names what it squashes — passes without a lease lookup, and a conflicted merge is judged only on the paths the merge changed, not on everything the base brought in. ddflow merge itself aborts a merge it fails, and doctor flags a primary checkout left mid-merge with the command that clears it. Over a hook the pre-commit framework generated, hooks install does not edit it (the next pre-commit install would discard the edit): it advises a repo: local hook in .pre-commit-config.yaml (ddflow precommit proposes it), says nothing needs installing when the framework already runs ddflow's check, and --force replaces the generated hook.
• Hooks find ddflow when they run, and fail open. The git hooks and the Claude/Gemini hook commands record the launcher that installed them, but try it only first: if that script or interpreter no longer exists or is no longer executable (a deleted venv, an uninstalled tool, a removed worktree) they run ddflow from PATH, and when there is none they print one line (ddflow: not found ... run: ddflow hooks install) and let the commit or turn go on. A missing tool never blocks a commit, but a check that silently stopped is reported: ddflow hooks status prints a DANGLING LAUNCHER line and ddflow doctor a problem (a note when ddflow on PATH still runs the hook) for each git hook, Claude/Gemini hook command and ddflow MCP entry (.mcp.json and the other agents' JSON configs) whose recorded path is gone, naming the refresh: ddflow hooks install (git hooks), ddflow hooks install --claude / --gemini (harness hooks), ddflow adopt (MCP entries).
• A per-task README check (decision D-readme-current). A task whose diff changed code under [enforce] readme_code_globs (default ddflow/**) but none of readme_files (default README.md), with no docs outcome recorded for it, is reported — README not updated: record the section you changed, or gate skip docs --reason ... — by ddflow complete, in ddflow gate status, and as a docs: line under the item in ddflow brief. Record the reason with ddflow gate skip <id> docs --reason "..." (or gate record <id> docs --outcome passed --evidence "<section>"). [enforce] readme_with_code picks the severity: warn (default), block (refuses complete) or off. Test files (tests/, test_*, *_test.*, *.test.*, *_spec.*), documentation (docs/, .md/.rst/.adoc/.txt) and ddflow's own event-log commits (.ddflow/**) are never reported; when git cannot say what the task changed, complete says the check could not run (a warning, never a block). It is a check, not a pipeline gate, so a task with nothing to document is not asked for a docs outcome; the phase-level docs gate is unchanged.
• A Claude Code SessionStart hook (ddflow hooks install --claude) puts the brief — crashed work to recover, ready items, binding decisions, operational memory — into every session, including after a context compaction, whether or not the agent remembers to ask. In a worktree behind its base branch it says so, and names the rulebooks that changed there. It is added beside the project's own hooks in .claude/settings.json, removed alone, and always exits 0.
• A prompt-capture hook (also installed by ddflow hooks install --claude, and by ddflow adopt for Claude Code and Gemini CLI) records every operator prompt without the agent having to remember session prompt. Claude Code's UserPromptSubmit hook runs ddflow hooks prompt with the hook JSON on stdin; it opens (once) and reuses a session keyed on the harness session id, redacts secrets before anything touches disk, appends one session.prompt, prints nothing, never waits on a silent stdin and always exits 0, so a failing ddflow never breaks the turn. An agent that also calls session prompt with the same words within two minutes is not recorded twice, and a hook that fires twice records once. Gemini CLI is wired the same way through its BeforeAgent hook (ddflow hooks install --gemini, .gemini/settings.json). Cursor (beforeSubmitPrompt in .cursor/hooks.json), Codex, Kilo and opencode have no wiring yet: known gaps. ddflow hooks status reports whether the prompt hook is installed, and ddflow hooks uninstall --claude (or --gemini) removes only ddflow's own hooks.
• Every prompt and note carries a session id, even an implicit one. The id on ddflow session prompt|note (and the MCP ddflow_session_prompt|note) is optional. Without one, the words go to the session named by DDFLOW_SESSION_ID/CLAUDE_SESSION_ID if that session is open, else this agent's most recently active open session, else a new session marked implicit in its start event. The output says which was used (recorded (0 redaction(s)) in session S (latest), ... (implicit, new)); an explicit id is unchanged and prints nothing extra. An ended session is never reused, text is never refused for a missing id (only empty text is), and redaction runs before disk as ever. The prompt hook does the same when a harness gives it no session id. Events recorded earlier with no session id show up as a ddflow doctor note; ddflow session adopt-orphans attaches each to the session nearest in time (writing a copy marked adopted_from, since the log is append-only; run twice it adds nothing, and replay shows the words once). With session.log_prompts = false a prompt is not recorded and the output says so, rather than opening a session for nothing.
• Sessions can be read back. ddflow session list prints sessions newest activity first with the agent, span, prompt/note/item counts, open or ended and an implicit marker (--state open|ended, --owner (the session's agent; --agent is who you are), --since <date|timestamp>, --limit, --json prints the same {rows, total, shown, truncated, filters} object as the other list commands; exit 2 when nothing matches, 3 for a bad filter). ddflow session show <id> prints one session: every prompt and note in order with its item, then the end summary, redacted like the exports; an unknown id is refused with the near matches. Both read the log directly, so they work with export sessions disabled; an adopted orphan appears once, and a prompt nobody has adopted yet belongs to no session (doctor counts those). No MCP tool yet (planned: ddflow_list kind=session).

Not automated, on purpose:

• Installing anything. Detection is read-only; the report is advice.
• The judgement inside an agent gate. ddflow records that you claim to have hunted bugs; it cannot check that you did. What it can do — and does — is make silence visible: a gate never run and never skipped blocks completion, so the failure mode is a refusal rather than a quiet omission.
• Deciding whether a plan is right. That is what a human = true gate is for.
• Pushing, releasing, or anything outward-facing.

The design assumption is that an agent's honesty cannot be verified, so the system is built to make an unverifiable claim expensive to make and easy to see: evidence contracts, mutation-verified gates, coverage gaps recorded on completion, and an exit code that distinguishes "could not" from "did not need to".

Testing

uv run pytest tests/ -q -n auto      # unit/integration tests, in parallel (pytest-xdist)
python3 demos/run_all.py             # 6 end-to-end scenarios, 219 assertions

While working on an item, ddflow tests --item <id> (MCP: ddflow_tests) lists the tests the change reaches — changed test files, tests importing a changed module directly or one step removed, tests named after a changed file, everything under a changed conftest.py — each with why, and prints one command that runs them in parallel with the project's own runner and worker flags. It is fast feedback and never a gate: the unit_tests gate always runs the whole suite, in parallel, because a targeted run hides breakage that was already there. The agent driver tells agents to run it after each change; ddflow workflow and ddflow doctor say when the configured test command uses one core.

The demos invent whole projects and drive them for real — real git worktrees, real pytest and npm test runs, real merges, real concurrent processes:

The scenarios and the stress test have found most of the bugs this project fixed; the unit tests found few of them. The full-lifecycle scenario was written to exercise the requirements rather than the code, and found four defects before it passed once — every one of them a CLI/MCP divergence that command-level parity could not see. The composed MCP run alone found eight that 213 unit tests and four other scenarios missed — including two that made core features useless out of the box. They all lived in seams: between two processes, between a read and a write, between two output surfaces, between a declared vocabulary and its callers, including one that does not reproduce below ~6 concurrent processes. They are catalogued with their regression tests in R6.

Documentation index