Skip to main content

AI Guider

A local co-pilot for AI coding agents.

AI Guider does not write your code. It sits beside agents (Cursor, Claude, Codex, and others) and helps them stay on track: clarify what you want, block out-of-scope work, and finish cleanly.

You → Cursor / Claude / Codex → AI Guider (local) → decisions, maps, and guardrails

Everything stays on your machine. No cloud account required for Guider itself.


What you get

Capability In plain terms
Missions A clear objective, success criteria, and open questions
Q&A gate The agent must ask you before guessing (stack, storage, etc.)
Plan refine Sharpens a plan from your answers — does not invent features
Scope checks Approve / caution / reject before risky changes
Codebase map Quick map of folders, entrypoints, and key symbols
Hooks (Cursor) Optional: block file edits until Guider approves the action

Install in 5 minutes

Requirements

  • macOS, Linux, or Windows
  • Python 3.10 or newer (3.12 recommended)

Step 1 — Install AI Guider

cd ~/ai-guider          # or wherever you cloned this repo
python3.12 -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate
pip install -e ".[dev]"

Check that it works:

ai-guider doctor

Step 2 — Connect your AI tools

Pick one command:

# Recommended: connect everything you use
ai-guider init --all-clients

# Or only what you need
ai-guider init --cursor --hooks     # Cursor + edit guardrails
ai-guider init --claude             # Claude Code
ai-guider init --claude-desktop     # Claude Desktop app
ai-guider init --codex              # OpenAI Codex

Step 3 — Restart the client

Tool What to do
Cursor Reload window (Cmd+Shift+P → “Developer: Reload Window”)
Claude Code Start a new session
Claude Desktop Quit completely, then reopen
Codex Restart; run codex mcp list and look for ai-guider

Full details: Installation guide · Connect Claude & Codex


How to use it (everyday flow)

You do not need to memorize tool names. Tell your agent:

Use AI Guider to govern this request:
Build a simple todo webapp, local only, no backend.

A well-behaved agent will roughly:

  1. Start a mission and ask you any missing questions
  2. Record your answers
  3. Refine and approve a plan
  4. Check scope before editing files
  5. Mark success criteria done, then complete

Step-by-step with examples: Usage guide


Useful commands

ai-guider --version       # Installed version
ai-guider help            # Full command guide (what each command does)
ai-guider doctor          # Is my install healthy?
ai-guider bootstrap       # Show active mission + tips for this session
ai-guider map             # Print a map of the current project
ai-guider resume          # What mission is active?
ai-guider missions        # List recent missions
ai-guider report          # Governance summary
ai-guider templates       # Mission templates (personal site, API, …)

Documentation

Browse all guides: docs/

Guide Audience
Installation First-time setup, troubleshooting
Usage Day-to-day workflow in plain language
Clients (Cursor, Claude, Codex) Per-app connection steps
Cursor deep dive Hooks, maps, Cursor-only features
Publishing to PyPI GitHub Release → PyPI workflow

License

MIT

Metadata

Release files for ai-guider 1.0.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for ai-guider 1.0.1
File Size Uploaded Provenance
ai_guider-1.0.1.tar.gz 79.9 kB With provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for ai-guider 1.0.1
File Interpreter ABI Platform
ai_guider-1.0.1-py3-none-any.whl Python 3 none any Details

Total release size: 159.7 kB

Release files / ai_guider-1.0.1.tar.gz

Download URL ai_guider-1.0.1.tar.gz
Size 79.9 kB
Tags Source
SHA-256 checksum
How to use checksums
ee098f3bb7f99cfa6d6cb5e720720c906508a6cf00196f87d6737efaab0d4b67
BLAKE2b-256 checksum
How to use checksums
257523d5c4eaf0cd19685c0d8f1f654c3a4c24a4518d06cfea4e742bb99765d7
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.12

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Jul 12, 2026.

Transparency log

Release files / ai_guider-1.0.1-py3-none-any.whl

Download URL ai_guider-1.0.1-py3-none-any.whl
Size 79.8 kB
Tags Python 3
SHA-256 checksum
How to use checksums
35f803be34aa9951b5525d04d78bf87cbfed7c16c9d5dcf4f2e869606b9fa045
BLAKE2b-256 checksum
How to use checksums
df5f38f5fa46a7a7b398b590140f511d27419319162ec0540b1f88127d24dd10
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.12

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Jul 12, 2026.

Transparency log

Security for ai-guider 1.0.1

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

All 2 files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for ai_guider-1.0.1.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Jul 12, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page