一键导入
research
Thoroughly investigate a part of the codebase before making changes. Use to understand existing patterns, trace data flows, and find reusable code.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Thoroughly investigate a part of the codebase before making changes. Use to understand existing patterns, trace data flows, and find reusable code.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Add a complete REST endpoint from command to controller route. Use when building new API endpoints.
Summarize the API interface changes made during THIS session for the frontend, then optionally spawn an agent to implement them in the frontend repo. Use after changing endpoints, request/response DTOs, or auth.
One-shot guided session — prunes unused code, scaffolds new entities/endpoints, updates project identity config, and replaces the init migration for a specific project.
Compress natural language memory files (CLAUDE.md, todos, preferences) into caveman format to save input tokens. Preserves all technical substance, code, URLs, and structure. Compressed version overwrites the original file. Human-readable backup saved as FILE.original.md. Trigger: /caveman-compress FILEPATH or "compress memory file"
Implement a change with a pipeline scaled to its complexity. Single entry point for all code changes — auto-detects effort or accepts small|mid|tuff override.
Run a Linear task end-to-end — creates an isolated tmux session + worktree, starts a Claude agent to implement it, opens a PR, and watches for review feedback automatically.
| name | research |
| description | Thoroughly investigate a part of the codebase before making changes. Use to understand existing patterns, trace data flows, and find reusable code. |
| argument-hint | <topic-or-area> |
Investigate a topic thoroughly. Read files, trace flows, find patterns. No code is written.
$0 -- Topic or area to research (e.g., "auth flow", "how events work", "user entity and its handlers")Find all files related to the topic:
find backend/domain/entities -name "*.py" — read relevant onesfind backend/domain/repos -name "*.py" — read Protocol interfacesfind backend/app/rest/v1/handlers -name "*.py" — read use casesfind backend/entry/rest/v1 -name "*.py" — read HTTP routesfind backend/app/shared/events -name "*.py" — read event definitionsfind backend/app/shared/ports -name "*.py" — read abstractionsfind backend/app/rest/v1/dtos -name "*.py" — read response shapesPick a representative endpoint and trace it from HTTP to database:
# Research: {topic}
## Existing Components
- {component}: {file path} — {what it does}
- ...
## Data Flow
{controller} -> {handler} -> {service?} -> {repo} -> {entity}
-> {event?} -> {event handler}
## Patterns to Reuse
- {pattern}: {file path} — {why it's relevant}
- ...
## Gaps
- {what's missing}: {where it should go}
- ...
## Recommendations
- {suggestion for implementation approach}