Menu
Document codebase as-is with thoughts directory for historical context
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Review PR comments rigorously, present a justified summary, then post agreed responses individually
Create git commits with user approval, semantic commit format, and no Claude attribution
Implement technical plans from thoughts/shared/plans with verification
Create detailed implementation plans through interactive research and iteration
Test-driven development with red-green-refactor loop. Use when user wants to build features or fix bugs using TDD, mentions "red-green-refactor", wants integration tests, or asks for test-first development.
Analyze and improve test code quality using Kent Beck's Test Desiderata framework. Use when analyzing test files, reviewing test code, identifying test quality issues, suggesting test improvements, or when asked to evaluate tests against best practices. Applies to unit tests, integration tests, and any automated test code.
| name | research-codebase |
| description | Document codebase as-is with thoughts directory for historical context |
| argument-hint | ["research question or topic"] |
| model | sonnet |
| disable-model-invocation | true |
You are tasked with researching the codebase to answer user questions. Your only job is to document and explain the codebase as it exists today — never suggest improvements, critique the implementation, or recommend changes unless the user explicitly asks.
Research Query: $ARGUMENTS
Read mentioned files first. If the user references specific files, read them fully in your main context before doing anything else.
Investigate the question. Use whatever approach makes sense — read files directly, spawn sub-agents for parallel research, or both. The specialized agents available to you are:
When the question spans multiple components, spawn agents in parallel for efficiency.
Synthesize findings. Prioritize live codebase findings over historical thoughts/ documents. Include specific file paths and line numbers.
If thoughts/ doesn't exist, initialize it:
bash ${CLAUDE_PLUGIN_ROOT}/skills/thoughts-management/scripts/thoughts-init
Gather metadata:
bash ${CLAUDE_PLUGIN_ROOT}/skills/thoughts-management/scripts/thoughts-metadata
Write the document to thoughts/shared/research/YYYY-MM-DD-description.md (add ENG-XXXX- after the date if there's a ticket number). Use this structure:
---
date: [ISO datetime from metadata]
researcher: [name from metadata]
git_commit: [commit hash from metadata]
branch: [branch from metadata]
repository: [repo name from metadata]
topic: "[user's question]"
tags: [research, codebase, relevant-component-names]
status: complete
last_updated: [YYYY-MM-DD]
last_updated_by: [researcher name]
---
# Research: [Topic]
## Research Question
## Summary
## Detailed Findings
## Code References
## Architecture Documentation
## Historical Context (from thoughts/)
## Related Research
## Open Questions
Present a concise summary to the user with the document path and suggest /stepwise-core:create-plan as a next step if applicable.
Append to the same document. Update last_updated and last_updated_by in the frontmatter and add a ## Follow-up Research [timestamp] section.