菜单
Deep research on a topic. Decomposes question into sub-questions, searches and fetches web sources in parallel, synthesizes a structured report with inline citations and a source list. Use when user asks for in-depth research, comprehensive analysis, or a structured report.
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
| name | deep-research |
| description | Deep research on a topic. Decomposes question into sub-questions, searches and fetches web sources in parallel, synthesizes a structured report with inline citations and a source list. Use when user asks for in-depth research, comprehensive analysis, or a structured report. |
| metadata | {"type":"feature","entry_action":"run","dependencies":"search,crawler,filesystem,deep-recall"} |
Use when the user asks for in-depth research, comprehensive analysis, or a structured report on a topic.
After run returns:
summary to the user immediately.report and MUST include the ## References section with all source URLs at the end.filesystem skill with tool write to save report to a file like research-<topic>.md.Do NOT auto-save. Always ask first.
Run deep research on a question. Returns a structured report, executive summary, key findings, and cited sources.
| Parameter | Type | Required | Description |
|---|---|---|---|
| question | string | Yes | The research question or topic |
| Field | Type | Description |
|---|---|---|
| report | string | Full markdown report with sections, inline citations, and a ## References section listing all cited URLs. |
| summary | string | 2-4 sentence executive summary. |
| sources | array | List of {url, title} objects — every URL referenced in the report, ordered by relevance. |
| sources_count | int | Total unique sources consulted across all sub-questions. |
| answer | string | Complete user-facing output (summary + report + references). When present, main agent passes it through directly without summarize. |
| error | string | Present only on failure; describes the error. |
Introduce Sophon and list available capabilities. Use ONLY when the user explicitly asks who you are, what you can do, what your capabilities are, or what skills you have. Do NOT call for greetings, small talk, or when the user has not asked about your identity or capabilities.
Process xlsx workbooks; fill columns from web via key column (URL→crawl, else search), LLM extract per batch. Tools: list_sheets, read_structure, read_sample, fill_by_column. Cross-sheet copy via copy_columns / copy_column_map.
System anomalies, errors, latency diagnosis. Logs, traces, metrics. Orchestrates log-analyze, trace, metrics primitives. Fetches data, frontend renders charts.
Web scraping via Playwright. Use when user wants to fetch page content or scrape a URL.
Memory recall and history exploration. Use for any question about past conversations, previous topics, what the user asked before, or recent activity. Replaces the memory skill.
Read and write Excel/CSV files. Supports .xlsx, .xls, and .csv formats.