fuyuxiang Deep Research
Multi-step research workflow that produces cited reports from multiple sources.
Workflow
- Decompose — Break the question into 3-5 sub-questions
- Search — Run multiple searches per sub-question (diverse keywords)
- Extract — Fetch and extract content from top URLs
- Cross-reference — Compare claims across sources, flag conflicts
- Synthesize — Write structured report with inline citations
Usage
python3 scripts/research_report.py "What are the best practices for LLM agent memory in 2026?"
python3 scripts/research_report.py "Compare FastAPI vs Litestar performance" --depth deep
Output Format
Reports follow this template:
# Research Report: {Topic}
**Date:** YYYY-MM-DD
**Sources consulted:** N
## Summary
2-3 sentence overview of findings.
## Key Findings
### Finding 1
Detail with evidence. [Source 1][1] confirms that...
### Finding 2
...
## Conflicting Information
Where sources disagree, note both positions.
## Conclusion
Actionable synthesis.
## Sources
[1]: https://... — Title
[2]: https://... — Title
Depth Levels
| Level | Searches | Pages Read | Time |
|---|---|---|---|
| quick | 2-3 | 3-5 | ~30s |
| normal | 5-8 | 8-12 | ~2min |
| deep | 10-15 | 15-25 | ~5min |
Combining with Other Skills
- Use
web-searchfor the search step - Use
web-extractfor the extraction step - Use
summarizeskill for per-page summaries before synthesis - Store results with
note-takingskill