メニュー
In-depth multi-step autonomous research — investigating complex questions, multi-source synthesis across web docs and codebase.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
After `nix flake update`, sync `input_rev` and `input_hash` metadata in all flake-input skill files to match new `flake.lock`.
Guides stable API and interface design. Use when designing APIs, module boundaries, or any public interface. Use when creating REST or GraphQL endpoints, defining type contracts between modules, or establishing boundaries between frontend and backend.
Tests in real browsers via Chrome DevTools MCP. Use when building or debugging anything that runs in a browser. Use when you need to inspect the DOM, capture console errors, analyze network requests, profile performance, or verify visual output with real runtime data. Requires the chrome-devtools MCP server to be configured.
Automates CI/CD pipeline setup. Use when setting up or modifying build and deployment pipelines. Use when you need to automate quality gates, configure test runners in CI, or establish deployment strategies.
Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensions before it enters the main branch.
Simplifies code for clarity. Use when refactoring code for clarity without changing behavior. Use when code works but is harder to read, maintain, or extend than it should be. Use when reviewing code that has accumulated unnecessary complexity.
| name | deep-research |
| description | In-depth multi-step autonomous research — investigating complex questions, multi-source synthesis across web docs and codebase. |
| metadata | {"when_to_use":"deep research, comprehensive report, thorough research, investigate, full picture, multi-source synthesis"} |
Conduct multi-step, autonomous research producing comprehensive, cited reports.
Identify knowledge gaps. Decompose into subtopics: which need web research, which need codebase exploration. Write the plan before launching subagents.
Break into 3-8 independent subtopics. Each: self-contained, narrow enough for one agent, mapped to web fetch or codebase search.
List 2-3 plausible but divergent answers before researching. Track which hypothesis evidence supports. Prevents anchoring.
Launch explore subagents per subtopic with: exact question, URLs to fetch, codebase areas to search, expected output format. Subagent instruction:
You are a deep-research subagent. Fetch relevant documentation from the web first.
Then search the codebase for how the thing is used in practice. Return: what it is,
why it's here, how it's wired, which hosts/modules use it. Cite paths and line numbers.
Identify gaps → focused follow-up agents. Resolve contradictions via primary docs. Note codebase vs docs discrepancies. Self-critique: what evidence would disprove the leading hypothesis? Alternative explanations? Confidence per finding?
Tag findings: High (verified in docs + codebase), Medium (one source), Low (inferred). Flag low-confidence dependencies as risks.
Aggregate into single report: grouped thematically, summary table, path/line citations, URL citations, architectural insights.
Run at least one verification command (test, lint, check) before presenting. Cross-check key claims.
flake.lock. Cross-reference.10 subtopics: do in waves of 5-8.
| Need | Tool |
|---|---|
| Project README / docs | webfetch |
| API reference / schema | webfetch |
| How X is used in codebase | grep |
| Where X is defined | glob |
| Read key files | read |
| Parallel exploration | task with explore |
| Aggregate findings | Direct synthesis |