Skip to content

README.md

Latest commit

 

History

History
57 lines (39 loc) · 2.58 KB

README.md

File metadata and controls

57 lines (39 loc) · 2.58 KB

Agent Collab Documentation

This folder contains the detailed maintainer and implementation documentation for Agent Collab. The root README.md is the public landing page. These files explain how the repository is organized, how the Python package works, how the CLI maps to .agent/ artifacts, and how the skill package should be maintained.

Reading Order

Document Use it for
Codebase Map A file-by-file tour of the repository.
Architecture The system model, package boundaries, and data flow.
CLI Reference Command behavior, options, defaults, and generated files.
Coordination Engine Internal implementation details for coordination.py.
Artifact Model The .agent/ directory, generated templates, naming rules, and ownership rules.
Skill Package How SKILL.md, references/, and agents/openai.yaml fit together.
Testing Unit test coverage, validation strategy, and smoke tests.
Development And Release Local setup, packaging, publishing, versioning, and maintenance workflow.
Maintenance Checklist Practical checklists for common repository changes.

Repository Layers

Agent Collab has four main layers:

  1. Public package layer

    pyproject.toml, README.md, LICENSE, and assets/ define how the project is presented, installed, licensed, and published.

  2. Agent skill layer

    SKILL.md, references/, and agents/openai.yaml define the agent-facing protocol and optional runtime metadata.

  3. Python package layer

    src/agent_collab/cli.py exposes the Typer CLI, while src/agent_collab/coordination.py creates .agent/ artifacts.

  4. Quality layer

    tests/ verifies helper behavior, CLI behavior, artifact paths, generated content, and entry point wiring.

Documentation Principles

  • Keep the root README.md fast to scan.
  • Put implementation and maintainer detail in docs/.
  • Keep agent-facing operational rules in SKILL.md and references/.
  • Keep docs aligned with generated artifact names and CLI defaults.
  • Treat docs as part of the product surface: changes to commands, templates, or workflow rules should update docs in the same change.

Quick Maintainer Commands

python3 -m pip install -e ".[dev]"
python3 -m pytest
python3 /path/to/quick_validate.py .
python3 -m build
python3 -m twine check dist/*

Use Development And Release for the full workflow.