菜单
Investigate two separate areas of the codebase simultaneously and combine findings into a comparison report. Use when you need research on two independent topics at once, want to compare two subsystems, or need to understand how two components relate to each other.
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
Explains code in plain language for someone unfamiliar with the programming language. Use when asked to explain code, walk through logic, describe what a function does, or when the user says "explain this" or "walk me through this".
Summarizes uncommitted git changes in a concise machine-readable format. Use in CI pipelines, scripts, or headless invocations where the output will be piped or captured.
Explains what a skill is and demonstrates that skills are working. Use when testing skills, when asked about skills, or when asked to demonstrate how skills work.
Demonstrates the personal scope for Claude Code skills. Use when testing personal-scoped skills or when the user wants to understand the difference between personal and project skill scopes.
Lists the conventions for this project and demonstrates the project scope for Claude Code skills. Use when asked about project conventions, code style, or as a demonstration of project-scoped skills.
Draft a CHANGELOG.md entry for the current changes in Keep a Changelog format. Use when releasing, tagging a version, or updating CHANGELOG.md.
| title | parallel-investigator |
| name | parallel-investigator |
| description | Investigate two separate areas of the codebase simultaneously and combine findings into a comparison report. Use when you need research on two independent topics at once, want to compare two subsystems, or need to understand how two components relate to each other. |
| disable-model-invocation | true |
| argument-hint | [topic-1] [topic-2] |
Investigate these two topics in parallel using separate Explore subagents:
Spawn two Explore subagents simultaneously — one per topic. Give each a self-contained research prompt that includes:
Do not run the subagents sequentially. Start both at the same time.
Wait for both subagents to return their results.
Combine the findings into a single comparison report with these sections:
Topic 1 findings — summary of what the first subagent found
Topic 2 findings — summary of what the second subagent found
Similarities — patterns, libraries, approaches, or design decisions shared between the two topics
Differences — how they diverge: in scale, in architecture, in data model, in error handling, or in how they are tested
Recommendation — if the user is trying to make a decision (e.g., which approach to follow, which subsystem to extend), give a direct recommendation with reasoning. If the topics are purely informational (e.g., understanding two independent systems), omit this section.
This skill does not use context: fork — it runs inline in your main conversation. This is intentional: a forked skill cannot spawn additional subagents, so any skill that needs to orchestrate multiple workers must run in the main context and delegate the worker tasks from there. The main conversation is the orchestrator; the Explore subagents are the workers.
