dadaia-workspace 0.1.5

AI-native workspace management CLI for multi-agent development with Claude Code, Codex and OpenCode

dadaia-workspace

AI-native workspace management for multi-agent development.

dadaia-workspace gives AI agents (Claude Code, Codex, OpenCode) a structured shared workspace with managed contexts, agentic asset projection, a real-time monitoring panel, and Spec-Driven Development enforcement.

dadaia-workspace is open source under the MIT license. Source code, issues, and contributions live at github.com/marcoaureliomenezes/dadaia-workspace.

---

Install

pip install dadaia-workspace

Requires Python 3.12+.

---

Quick start

dadaia init        # initialise .dadaia/ and project agentic assets into .claude/
dadaia doctor      # health check: contexts, assets, panel, SDD gate
dadaia panel       # launch the local dashboard at http://localhost:8080

That's it. From here, agents self-discover the rest via dadaia --help.

---

CLI reference

dadaia [COMMAND] --help   # always works at every level

Command group	What it does
dadaia init	Bootstrap workspace: creates .dadaia/, projects agent assets
dadaia doctor [--fix]	Diagnose and repair workspace state
dadaia context	Manage Spec Context Projects (activate, deactivate, promote, switch)
dadaia repos	Query the known repos catalog
dadaia public	Stage and install distributed agentic assets across runtimes
dadaia server	Dev server port registry (register, list, deregister)
dadaia panel	Start local monitoring panel (http://localhost:8080)
dadaia specs	SDD release-lifecycle checks and helpers
dadaia reports	Inspect and validate agent handoff reports
dadaia orchestrate	Run multi-agent workflows
dadaia academy	Manage Academy courses
dadaia export / import	Portable workspace archive

Context management

dadaia context list                  # list all Spec Context Projects
dadaia context activate <name>       # activate a context (clones repo if absent)
dadaia context show                  # show active context
dadaia context show --json           # machine-readable active context (agent use)
dadaia context promote <name>        # set as workspace primary
dadaia context deactivate <name>     # git-sync and remove from disk
dadaia context use <name>            # isolate shell session to a context

Asset pipeline

dadaia public stage                          # stage assets from package into .dadaia/agentic/
dadaia public install                        # project assets to all AI runtimes
dadaia public install --target claude        # project to Claude Code only
dadaia public install --target codex         # project to Codex only
dadaia public doctor                         # detect drift between source → staging → projection

Dev server registry

dadaia server register <port> <project> <url>   # register a running dev server
dadaia server list                               # list all registered servers
dadaia server deregister <port>                  # remove from registry

---

Agent self-discovery protocol

Agents can fully discover and operate dadaia-workspace with no prior knowledge:

1.  dadaia --help                    → full command tree
2.  dadaia doctor                    → workspace health (what's wrong, what to fix)
3.  dadaia context show --json       → active Spec Context Project (machine-readable)
4.  dadaia public doctor             → agentic asset projection state
5.  GET http://localhost:8080/health → panel health probe (returns {"status":"ok","version":"..."})

If dadaia doctor exits non-zero, run dadaia doctor --fix before doing anything else.

---

What is a Spec Context Project?

A Spec Context Project is a scoped workspace unit: a git repo with a specs/ directory following the SDD (Spec-Driven Development) lifecycle — SPEC.md → PLAN.md → TASKS.md → CLOSURE.md. The dadaia context commands let you activate, switch, and isolate contexts across projects.

---

Supported AI runtimes

Runtime	Projection target
Claude Code	.claude/agents/, .claude/rules/, .claude/skills/, .claude/workflows/
Codex	.codex/agents/*.toml, .codex/rules/, .codex/skills/
OpenCode	.opencode/agents/, .opencode/skills/, .opencode/workflows/
Generic agents	.agents/skills/, .agents/workflows/

All projections are generated from a single canonical source at dadaia_workspace/public/ via dadaia public stage && dadaia public install.

---

Panel health probe

The panel exposes a public (no auth) health endpoint for agent monitoring:

GET http://localhost:8080/health
→ {"status": "ok", "version": "0.1.2"}

Use this in automated checks to verify the panel is running before making authenticated API calls.

---

SDD gate

dadaia-workspace ships a pre-tool hook (sdd-spec-gate.sh) that enforces Spec-Driven Development: agents cannot edit production files without an active task marker ([-]) in TASKS.md. The gate is installed automatically by dadaia init and dadaia public install.

---

Links

• PyPI: https://pypi.org/project/dadaia-workspace/
• GitHub: https://github.com/marcoaureliomenezes/dadaia-workspace
• Issues: https://github.com/marcoaureliomenezes/dadaia-workspace/issues