ワンクリックで
web-research
Use this skill for requests related to web research; it provides a structured approach to conducting comprehensive web research
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。
メニュー
Use this skill for requests related to web research; it provides a structured approach to conducting comprehensive web research
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
Load when the MainAgent is assembling a comparison report across multiple narrative-direction explorations (results from parallel/sequential task(explorer) calls). Use to compare directions along narrative, character, cost, and reader-promise dimensions. "Branches" here means narrative branches — this is NOT a git branch tool.
Plan and track a character's long-term transformation arc, ensuring each action reflects their current stage of change.
Check physical, social, and psychological plausibility in fiction scenes.
Load when WriterAgent finishes a chapter draft or is planning a chapter that should set up later payoffs. Use to embed plant-able details (objects, gestures, line drops) for ch+2 or later, and to verify earlier plants get acknowledged. Skip for the very first chapter (no prior plants exist) and for pure dialogue scenes with no narrative scaffolding.
Plan or draft scenes where rhythm, tension, and momentum matter by balancing escalation and release.
Plan or draft a fiction scene with a concrete character goal, meaningful opposition, a turning point, and a consequential result.
| name | web-research |
| description | Use this skill for requests related to web research; it provides a structured approach to conducting comprehensive web research |
This skill provides a structured approach to conducting comprehensive web research using the task tool to spawn research subagents. It emphasizes planning, efficient delegation, and systematic synthesis of findings.
Use this skill when you need to:
Scope note: This is a general research workflow. For named-author writing style extraction, use the writing-style skill to orchestrate Researcher, WritingStyleExtractor, and WritingStyleSkillCreator.
Before delegating to subagents, you MUST:
Create a research folder - Organize all research files in a dedicated folder under the internal virtual filesystem, using an absolute path:
mkdir /large_tool_results/research_[topic_name]
This keeps files organized, prevents clutter in the user's workspace, and avoids requiring file-write approval.
Analyze the research question - Break it down into distinct, non-overlapping subtopics
Write a research plan file - Use the write_file tool to create /large_tool_results/research_[topic_name]/research_plan.md containing:
Planning Guidelines:
For each subtopic in your plan:
Use the task tool to spawn a research subagent with:
/large_tool_results/research_[topic_name]/findings_[subtopic].mdRun up to 3 subagents in parallel for efficient research
Subagent Instructions Template:
Research [SPECIFIC TOPIC]. Use the web_search tool to gather information.
After completing your research, use write_file to save your findings to /large_tool_results/research_[topic_name]/findings_[subtopic].md.
Include key facts, relevant quotes, and source URLs.
Use 3-5 web searches maximum.
After all subagents complete:
Review the findings files that were saved:
ls(path="/large_tool_results/research_[topic_name]") to see what files were createdread_file with the file paths (e.g., /large_tool_results/research_[topic_name]/findings_*.md)read_file for LOCAL files only, not URLsSynthesize the information - Create a comprehensive response that:
Write final report (optional) - Use write_file to create /large_tool_results/research_[topic_name]/research_report.md if requested
Note: If you need to fetch additional information from URLs, use the fetch_url tool, not read_file.
You have access to:
Each subagent you spawn will have access to: