Codebase Mentor
Codebase Wizard is a Claude Code plugin that explains any codebase, spec, or markdown artifact through a wizard-style conversational Q&A interface. Every answer is anchored to actual code shown inline. Sessions are auto-captured and synthesizable as structured documentation.
What It Does
- Describe mode — walks a repo owner through Q&A and produces
CODEBASE.md - Explore mode — gives a new developer a learning-order tour and produces
TOUR.md - File mode — walks through any markdown file section-by-section
- Auto-capture — every session is captured to JSON via Agent Rulez hooks
- Export — synthesize raw session logs into structured docs on demand
Install
pip install ai-codebase-mentor
ai-codebase-mentor install --for claude
This copies the Codebase Wizard plugin to ~/.claude/plugins/codebase-wizard/.
For a per-project install:
ai-codebase-mentor install --for claude --project
Usage
After installing, in any Claude Code session:
/codebase-wizard-setup # one-time setup: installs hooks and writes permissions
/codebase-wizard # start a wizard session
/codebase-wizard-export # synthesize captured sessions into docs
Uninstall
ai-codebase-mentor uninstall --for claude
Status
ai-codebase-mentor status
CLI Reference
ai-codebase-mentor install --for [claude|all] [--project]
ai-codebase-mentor uninstall --for [claude|all] [--project]
ai-codebase-mentor status
ai-codebase-mentor version
Commands
| Command | Description |
|---|---|
/codebase-wizard |
Start a wizard session (Describe, Explore, or Ask mode) |
/codebase-wizard-setup |
One-time onboarding: install Agent Rulez hooks, write scoped permissions |
/codebase-wizard-export |
Synthesize captured session JSON into CODEBASE.md, TOUR.md, or FILE-NOTES.md |
How It Works
The plugin follows a 5-phase conversational flow:
| Phase | Trigger | What loads |
|---|---|---|
| 1 — Repo Scan | User shares repo | scan-patterns.md |
| 2 — Question Handling | "how does X work?" | (none extra) |
| 3 — Navigation | "rewind", "jump to" | navigation-commands.md |
| 4 — Tutorial Mode | README found, or "teach me" | tutorial-mode.md |
| 5 — Persistence | "save", "export" | persistence.md |
Every answer follows the same loop:
- Find the relevant code
- Show it as a code block with a full anchor (
src/auth/middleware.ts → validate() → L14-31) - Explain in plain English
- Show connections (calls / called-by)
- Predict 2–3 follow-up options
Milestones
| Version | Runtime | Status |
|---|---|---|
| v1.0 | Claude Code | Complete |
| v1.2 | OpenCode + PyPI publish | Planned |
| v1.3 | Codex (subagent-aware) | Planned |
| v1.4 | Gemini CLI | Planned |
| v1.5 | LangChain DeepAgent standalone | Planned |
Development
git clone https://github.com/SpillwaveSolutions/codebase-mentor
cd codebase-mentor
pip install -e .
pytest tests/
License
MIT — Spillwave Solutions
Metadata
Release files for ai-codebase-mentor 1.3.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| ai_codebase_mentor-1.3.0.tar.gz | 64.1 kB | With provenance | Details for ai_codebase_mentor-1.3.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| ai_codebase_mentor-1.3.0-py3-none-any.whl | Python 3 | none | any | Details for ai_codebase_mentor-1.3.0-py3-none-any.whl |
Total release size: 126.4 kB
Release files / ai_codebase_mentor-1.3.0.tar.gz
| Download URL | ai_codebase_mentor-1.3.0.tar.gz |
|---|---|
| Size | 64.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
ccc60080e3f75c2192ffe4d64d538e99915a3e30d9af477d69e5aeb8bd9aa9ca
|
|
BLAKE2b-256 checksum How to use checksums |
bc5aef0d230d461def5fbe1f07206e9c03dbb901e774f632a47b94c76ccdbfb1
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
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 Mar 31, 2026.
Transparency logRelease files / ai_codebase_mentor-1.3.0-py3-none-any.whl
| Download URL | ai_codebase_mentor-1.3.0-py3-none-any.whl |
|---|---|
| Size | 62.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
83dbc6944b8c579fb75ef1f5ea3144d80b489b4c812280fbb6f5c2af348ea1d2
|
|
BLAKE2b-256 checksum How to use checksums |
92e10281c706f859ed4d820ad556a0138ccceea37ca8b96b284387530c3cafdd
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.7
|
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 Mar 31, 2026.
Transparency logSecurity for ai-codebase-mentor 1.3.0
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.
Attestation for ai_codebase_mentor-1.3.0.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 Mar 31, 2026.
Transparency logA 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.