원클릭으로
researching-codebase
Investigates codebase before planning. Use before any non-trivial implementation task to gather context in isolation.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Investigates codebase before planning. Use before any non-trivial implementation task to gather context in isolation.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Compacts verbose context into structured summary. Use after pollution sources (searches, logs, JSON) or at phase milestones.
Create a pull request from the current branch. Analyzes commits, generates title+body from PR template, pauses for approval, then pushes and creates PR. Use after committing changes.
Designs concise, streamlined backend systems matching exact task requirements. Use when planning APIs, data models, system architecture, or when the user requests backend design work.
Audits and aligns project documentation against authority chains (project docs and Claude Code infrastructure). Detects broken references, duplicates, scope creep, and chain breaks. Use when reviewing documentation health, fixing stale references, or enforcing single-source-of-truth.
Interactive Q&A to build UserStory.md from user input. Use when the user wants to create a user story document or start the assisted workflow.
Generates prd.json task tracking file from PRD.md requirements document. Use when initializing Ralph loop or when the user asks to convert PRD to JSON format for autonomous execution.
| name | researching-codebase |
| description | Investigates codebase before planning. Use before any non-trivial implementation task to gather context in isolation. |
| compatibility | Designed for Claude Code |
| metadata | {"argument-hint":["topic-or-question"],"context":"fork","agent":"Explore","allowed-tools":"Read, Grep, Glob"} |
Query: $ARGUMENTS
Gathers codebase context in isolation before planning. Prevents search artifacts from polluting main context.
Follow ACE-FCA quality equation: Correct + Complete + Minimal noise
---
research_query: "<original question>"
timestamp: "<ISO 8601>"
files_examined: <count>
---
## Key Files
| File | Purpose | Key Lines |
|------|---------|-----------|
| `path/to/file.ext` | Brief purpose | L42-58 |
## Patterns
- **Pattern name**: Description with file reference (`path:line`)
## Constraints
- Constraint with evidence (`path:line`)
Every claim must include:
src/auth/login.ts)L42-58 or L127)Bad: "Authentication uses JWT tokens"
Good: "Authentication uses JWT tokens (src/auth/jwt.ts:L23-45, verifyToken function)"
See .claude/rules/context-management.md and .claude/rules/core-principles.md.