Skip to main content

Scaffold agent context files into a project.

Project description

agentinit

CI PyPI Python Versions

agentinit preview

Scaffold and maintain agent context files for modern coding assistants, with a deterministic, standard-library-only CLI.

agentinit creates a clean router-first setup around AGENTS.md, plus companion files for Claude, Cursor, Copilot, Gemini CLI, and llms.txt.

Why agentinit 🎯

  • Single source of truth: keep high-level rules in AGENTS.md.
  • Low drift workflow: regenerate and verify router files with sync --check.
  • Lean context model: keep short entry files and push detail into docs/.
  • No runtime dependencies: pure Python stdlib.

Quick Start 🚀

Requires Python 3.10+.

# Install (recommended)
pipx install agentinit

# Initialize in an existing repository
cd your-project
agentinit init --minimal

Minimal profile generates:

your-project/
├── llms.txt
├── AGENTS.md
├── CLAUDE.md
└── docs/
    ├── PROJECT.md
    └── CONVENTIONS.md

Full profile (agentinit init) also includes GEMINI.md, docs/STATE.md, docs/TODO.md, docs/DECISIONS.md, Cursor/Copilot/Claude rule files, and .contextlintrc.json.

Core Workflow 🧭

# 1) Bootstrap context files
agentinit init --detect --purpose "AI code review assistant"

# 2) Keep llms.txt aligned with project docs
agentinit refresh-llms

# 3) Add modular resources
agentinit add skill code-reviewer
agentinit add mcp github
agentinit add security

# 4) Validate quality gates
agentinit status --check
agentinit sync --check
agentinit lint

For minimal projects, both status --check and sync --check auto-detect the generated minimal profile. status --minimal --check and sync --minimal --check remain available if you want to force that mode explicitly.

Command Reference

  • agentinit init add missing context files in current directory
  • agentinit minimal shortcut for init --minimal
  • agentinit new <project> create a new project and scaffold context
  • agentinit refresh-llms (alias: refresh) regenerate llms.txt
  • agentinit sync reconcile router files from templates
  • agentinit status show missing/incomplete files and line budgets
  • agentinit lint run contextlint checks
  • agentinit add <type> <name> install resources (skill, mcp, security, soul)
  • agentinit remove remove or archive managed files

CI Example ✅

Use both structure and drift checks:

agentinit sync --check
agentinit status --check
agentinit lint

Tool Compatibility 🤝

agentinit is designed to work with common agentic workflows by generating:

  • AGENTS.md as primary router
  • CLAUDE.md for Claude Code memory/routing
  • .cursor/rules/project.mdc for Cursor rule routing
  • .github/copilot-instructions.md for GitHub Copilot context
  • GEMINI.md for Gemini CLI context routing
  • llms.txt as project discovery index

Troubleshooting 🛠️

If your agent cannot find context files:

  • track files in git (git add AGENTS.md CLAUDE.md GEMINI.md llms.txt docs/)
  • verify ignored files (git status --ignored)
  • regenerate derived files (agentinit refresh-llms and agentinit sync)
  • replace managed symlinks with regular files inside the repo; unsafe managed paths are skipped by design

Documentation 📚

Wiki (full usage and examples):

Development 🧪

pip install -e . --group dev
python3 -m ruff check agentinit tests cli
python3 -m ruff format --check agentinit tests cli
python3 -m pytest tests/ -v

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

agentinit-0.3.8.tar.gz (60.9 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

agentinit-0.3.8-py3-none-any.whl (59.7 kB view details)

Uploaded Python 3

File details

Details for the file agentinit-0.3.8.tar.gz.

File metadata

  • Download URL: agentinit-0.3.8.tar.gz
  • Upload date:
  • Size: 60.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for agentinit-0.3.8.tar.gz
Algorithm Hash digest
SHA256 c8e3a2a882c17ff67b95fed7cd66fe360cc50c782e453bb2f521eb7b1716e954
MD5 07408e30a0d166f138a867c969fc4573
BLAKE2b-256 915f2bd0f1987ac83d28437b71e384b99ec3286618e4d071579d4ae0d05cebf0

See more details on using hashes here.

Provenance

The following attestation bundles were made for agentinit-0.3.8.tar.gz:

Publisher: publish.yml on Lucenx9/agentinit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file agentinit-0.3.8-py3-none-any.whl.

File metadata

  • Download URL: agentinit-0.3.8-py3-none-any.whl
  • Upload date:
  • Size: 59.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for agentinit-0.3.8-py3-none-any.whl
Algorithm Hash digest
SHA256 89d064843eb3a7e9e6f3973ef554d57402ede1b2c7cca9cb8dc9c112440fda0c
MD5 bba2084dfb37ed02fc6f43cc87864fb5
BLAKE2b-256 6ac2af453ad561d9a8650e690443fae54a54c27dea28725cb83170a96f89074d

See more details on using hashes here.

Provenance

The following attestation bundles were made for agentinit-0.3.8-py3-none-any.whl:

Publisher: publish.yml on Lucenx9/agentinit

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page