reactifact 0.11.1

Event-driven agents for Python developers — tasks wake on typed artifacts, like Celery tasks wake on messages. No graph to draw.

Python developers already know this model from Celery: define a task, declare what triggers it, let the runtime run it. reactifact applies it to agents — a task reacts to a typed, versioned artifact appearing in the context, not to a queue message you push or a graph edge you draw. The runtime derives what runs next from state.

Single process today (no broker, no worker pool) — the model is Celery-shaped, not its distributed runtime.

On top of that model you get something a task queue doesn't: every artifact is versioned with provenance, so a run is reproducible (context_hash) and auditable (audit.report) for free. The model reasons; the arithmetic stays deterministic; every claim carries provenance.

On top: a provable answer

examples/fintech_audit — a transactions CSV, a budget CSV and a policy doc, no API key. The model never produces the number; plain Python does, and the answer is linked to its evidence:

$ .venv/bin/python -m examples.fintech_audit.main

  cloud spend:        $45,000   (2026-04 $12k, 2026-05 $15k, 2026-06 $18k)
  variance vs budget: +12.5%    budget $40,000, policy threshold 10% → over

  answer:    Q2 cloud spend was $45,000 against a $40,000 budget (+12.5%) —
             exceeds the 10% policy threshold. CFO approval is required.
  citations: budget.csv, transactions.csv, policy.md
  audit:     Answer —supported_by→ {Variance, Spend, Table, Policy}
             answer sha256 5461290d…  ·  context sha256 24449f6f…

  >>> re-running the pipeline hashes identically — or verify a saved run:
  >>> reactifact replay <store> --session <id> --verify 24449f6f…

That hash is the whole point: the answer is reproducible and its provenance is a queryable graph (context.related(answer.id, "supported_by")), not a log line.

The number is computed, and its provenance recorded, in plain Python — the model is never the source of truth (trimmed from examples/fintech_audit/produce.py):

@produce(Variance, reacts_to=Spend)              # wakes when a Spend artifact exists
async def compute_variance(call: ProduceCall) -> None:
    spend = call.trigger                         # the artifact that triggered this run
    budget = cloud_budget(call.context)          # read from budget.csv
    pct = (spend.data.total - budget) / budget   # deterministic — never the model
    variance = call.effects.create_once_from(
        spend,                                   # stable id → idempotent re-runs (§42)
        Variance(actual=spend.data.total, budget=budget, pct=pct,
                 threshold=0.10, within_policy=abs(pct) <= 0.10),
    )
    variance.link("calculated_from", spend)      # provenance edge, queryable


answer = ctx.latest(AuditAnswer)
print(report_to_markdown(build_report(ctx, answer)))  # hash per artifact + edges
print(context_hash(ctx))                              # reproducible fingerprint

Two agents explore independently on their own forks and merge back automatically — and when they disagree, reactifact refuses to merge silently:

pip install reactifact

Runs offline, no API key needed — paste this straight into a .py file. Two agents, no graph edge declared between them — the second reacts because the first one's output exists, and the answer carries proof of where it came from:

from pydantic import BaseModel

from reactifact import Budget, Consume, Context, Runtime, RuntimeResources, create_agent, produce


class Question(BaseModel):
    text: str


class Evidence(BaseModel):
    text: str


class Answer(BaseModel):
    text: str


DOCS = {
    "refund": "Refunds are available within 14 days of purchase.",
    "pricing": "The Pro plan is $49/month, billed annually.",
}


@produce(Evidence)
async def find_evidence(call):
    question = next((a for a in call.inputs if isinstance(a.data, Question)), None)
    if question is None:
        return None
    hit = next((v for k, v in DOCS.items() if k in question.data.text.lower()), None)
    if hit is not None:
        call.effects.create(Evidence(text=hit))


@produce(Answer)
async def answer_from_evidence(call):
    evidence = next((a for a in call.inputs if isinstance(a.data, Evidence)), None)
    if evidence is None:
        return None
    call.effects.create(Answer(text=evidence.data.text)).link("supported_by", evidence)


search_agent = create_agent("search", consumes=[Consume(Question)], produces=[find_evidence])
answer_agent = create_agent("answer", consumes=[Consume(Evidence)], produces=[answer_from_evidence])

ctx = Context(resources=RuntimeResources())
runtime = Runtime(ctx, agents=[search_agent, answer_agent], budget=Budget(max_runs=10))

ctx.create(Question(text="what's your refund policy?"))
runtime.run()  # search_agent and answer_agent both react — nobody wired them together

answer = ctx.latest(Answer)
evidence = ctx.related(answer.id, "supported_by")[0]
print(answer.data.text)                     # "Refunds are available within 14 days of purchase."
print("supported_by:", evidence.data.text)  # provenance you can trace, not just a string in a log

The same idea, live — the knowledge example's CLI answering a harder, multi-source question (docs + a CSV) with a real computed number and its sources, no LLM key required:

How it works

ARTIFACT CREATED / UPDATED
       │
       ▼
     AGENTS REACT ──self.effects──► Effects ──compile──► Patch
       ▲                                                      │
       └──────────────────────────────────────────────────────┘
                                                        Context v+1

A produce writes what should change (self.effects.create/update/link/ask) and returns None; the runtime compiles the effect set into one atomic Patch and moves the context to the next version. The event that wakes an agent is derived from that same change — the causal chain can never drift from the actual state.

What makes it different

Reactive. Deterministic. Accountable.

Full breakdown, including where reactifact is not the right choice: docs/en/comparison.md.

Proof, not a claim — examples/ledger is a 4-artifact billing calc (LaborCost, Tax, Discount, Total) with no LLM, fully offline. Edit one fact and see what actually reruns:

>>> editing ONLY TaxRate (0.08 -> 0.12) — a fact nothing about
>>> LaborCost or Discount ever consumed.

  LaborCost    value=500.0      version=0   # untouched
  Tax          value=60.0       version=1   # recomputed
  Discount     value=25.0       version=0   # untouched
  Total        value=535.0      version=1   # recomputed

2 of 4 artifacts recompute — the 2 that actually depend on TaxRate — because Artifact.version tracks real consumption, not a graph edge you drew by hand. Run it yourself: uv run python -m examples.ledger.main.

Core primitives

• Context — versioned working state, git-like commits, diff/rollback/merge.
• Artifact — a first-class typed object (Claim, Evidence, Answer), not a string blob.
• Effects — an agent states its change via self.effects.create/update/link/ask; the runtime compiles it.
• Patch — the compiled, validated change-set applied as one atomic commit.
• Agent — a thin container declaring consumes/produces; logic lives in a Produce.
• Source — retrieval is a capability: vector search is one strategy, not the only one; filesystem, CSV, and the web are equally first-class today (direct API, keyword, and SQL sources are on the roadmap, not yet shipped).
• Provenance — every derived artifact links to what produced it (Answer —supported_by→ Claim —derived_from→ Evidence —extracted_from→ Doc).
• HITL — humans as effects.ask(...) → PendingQuestion, answered via effects.resume(...) like any agent.

In the box

• Deterministic by design — calculations over structured data, honest None fallbacks instead of hallucinated answers; the model reasons, never "knows".
• Observability — every run traces agent spans, reads/writes, LLM calls, tokens: SQLite store + web dashboard, exportable to Langfuse/Postgres (async sinks).
• MCP, both ways — call any MCP server's tools as a Tool (mcp_stdio_tools/mcp_http_tools), or expose your own Tools and a live Context as an MCP server (create_mcp_server) for Claude Desktop, Claude Code, or another agent to call into (mcp extra).
• Budgets & replanning — cap by runs/time/iterations/tool-calls, replan on decline.
• Branching & replay — context.branch(), three-way merge(), deterministic ReplayLLM, all for audit and safe alternative states.
• Sessions — SessionStore over FileKVBackend/SQLiteKVBackend (and PostgreSQLKVBackend) for durable chat memory.
• Web layer — ChatAssistant + create_chat_router mount a canonical SSE chat on your FastAPI app; errors degrade to a logged fallback, never a 500.
• Recipes — find/find_all (typed lookup in inputs), fan_out_sources, materialize_doc, StatusMachine, WindowSummarizer/WindowPruner (bounded conversation memory), change→rebuild rollback helpers, Skill/match_skills (Claude-Skills-shaped instructions, keyword-triggered) — pure and LLM-free, except the summarizer, which takes your callback.
• Viz & CLI — Mermaid blueprint/context_to_mermaid/trace_to_mermaid; reactifact with graph/context/trace/replay/branch.

Run a demo

Offline-capable, no API keys required (deterministic fallbacks):

uv run python ./examples/llm_ladder/level1.py  # the simplest LLM turn (offline too)
uv run python ./examples/repair/web.py         # room renovation: plan, estimate, CSV export
uv run python ./examples/devops/web.py         # HITL ops assistant + trace dashboard

Classic-pattern ports run as one-liners too: python -m examples.{reflection,map_reduce,supervisor,summarize,time_travel,plan_execute,adaptive,ledger}.main.

Prefer a notebook? — examples/quickstart.ipynb covers the four quick cases (agent/rag/tools_agent/chat_agent), offline by default.

Examples (in-repo, not shipped)

• knowledge — multi-source chat: search → evidence → claim verification → answer, with CSV calculation.
• research — goes to the web (WebSource): lazy page fetch → evidence → verified claims → answer with URL provenance.
• medic-lab — hypothesis laboratory: competing hypotheses scored, HITL steering, honest report.
• devops — HITL tool agents + LLM tool router + trace dashboard.
• repair — budget-aware replanning (chat/data in Russian by design).
• forklab — deterministic branch & merge: two strategies on their own forks, three-way merge.
• ledger — offline proof of reactive recompute: edit one fact, only its real Consumers re-run.
• llm_ladder — the workflow from one LLM call to state-changing patches (3 levels).
• adaptive — hybrid scheduler: rule filters + deterministic rank + LLM tie-break + rank_limit.
• {reflection,map_reduce,supervisor,summarize,time_travel,plan_execute} — canonical ports (see port-matrix).

Documentation

• English · Русский — concepts, sources, providers, recipes, patterns, observability, eval, branching, replay, viz/CLI, API.
• Quickstart — the four quick cases (structured call, RAG, tools, chat) in a few lines; also as a Colab notebook.
• Migrating — from LangGraph / CrewAI / LlamaIndex / plain Python: concept map, port-one-node, interop, checklist.
• Why reactifact — the design argument: why effects, why no graph, why determinism.
• Comparison — reactifact vs LangGraph/CrewAI, feature by feature, and when not to use reactifact.
• Tutorial · llm-ladder — learn the workflow.
• docs/constitution.md — the full design rationale and invariants.
• Roadmap — what's next, and what's deliberately out of scope.

Development

uv sync --extra dev --extra web
.venv/bin/python -m pytest
.venv/bin/mypy
.venv/bin/ruff check

License

MIT — see LICENSE.