staruhub/ClaudeSkills / skills/Geek-skills-deep-research
deep-research
VerifiedUse this skill when the user wants an evidence-based research memo, literature review, market/policy/technical landscape, or a multi-source decision brief with citations, trade-offs, and a clear conclusion. Best for tasks that need synthesis across multiple external sources, iterative follow-up research, or a reusable written artifact. Do not use for quick factual lookups, single-source summaries, simple Q&A, summarizing one document the user already provided, plan-only requests where the user explicitly defers the actual research, or when the user wants a short answer with no need for cited evidence (a short but evidence-backed conclusion is still a brief memo, which this skill covers). Chinese trigger examples: "帮我调研", "深度研究", "综述报告", "技术选型分析", "竞品研究", "政策分析".
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 11.0 KB |
| assets/report_template.md | markdown | 1006 B |
| evals/routing-evals.json | config | 3.8 KB |
| evals/runbook.md | markdown | 967 B |
| references/evaluator-prompt.md | markdown | 4.0 KB |
| references/handoff-format.md | markdown | 3.9 KB |
| references/landscape-scan.md | markdown | 1.2 KB |
| references/methodology.md | markdown | 5.7 KB |
| references/observability.md | markdown | 3.0 KB |
| references/quality-gates.md | markdown | 5.5 KB |
| references/report-assembly.md | markdown | 3.4 KB |
| references/research-notes-format.md | markdown | 2.2 KB |
| references/subagent-prompt.md | markdown | 1.7 KB |
| references/tension-discovery.md | markdown | 1.2 KB |
| scripts/emit_run_summary.py | script | 5.0 KB |
| scripts/source_evaluator.py | script | 11.9 KB |
| scripts/verify_citations.py | script | 10.7 KB |
references/report-assembly.md
View on GitHub# Report Assembly V8
## Core rule
Write the final artifact from:
- the research plan
- the notes
- the approved registry
- raw/fetched sources only when you need to resolve a critical conflict or verify a high-stakes claim
The goal is **notes-first synthesis**, not blind copy-paste from search results.
---
## Pick the right artifact shape
### A. Brief memo
Use when the user wants a compact but defensible answer.
Suggested structure:
1. Direct answer / recommendation
2. Key evidence
3. Trade-offs / limitations
4. Optional next step
### B. Full report
Use when the user asked for comprehensive analysis.
Suggested structure:
1. Executive summary
2. Background / framing
3. Main analysis sections
4. Limitations & trade-offs
5. Decision framework (if needed)
6. References
### C. Delta update
Use when continuing a prior round.
Suggested structure:
1. What changed
2. New evidence
3. Impact on prior conclusion
4. Remaining uncertainty
---
## Mandatory sections / elements
Every final artifact must contain:
- a clear answer to the question
- source-backed findings
- limitations / trade-offs
- calibrated uncertainty
- references or inline source mapping
### Decision framework
Include only when the user is making a choice.
Do not force it into descriptive overviews.
### Non-obvious insight
Every good artifact should surface at least one useful nuance.
For contested or hype-heavy topics, this may be an assumption the evidence challenged.
For stable topics, this can simply be the most decision-relevant nuance the user might otherwise miss.
---
## Writing order
1. Body / main analysis
2. Limitations & trade-offs
3. Decision framework (if needed)
4. Executive summary / opening answer
5. References
Do **not** write the summary first.
---
## Evidence and wording rules
- Every important factual claim should be cited
- Every important number should be cited
- If the evidence is mixed, say it is mixed
- Do not turn thin evidence into precise certainty
- If you infer, label it as your analysis
### Good phrasing
- “The evidence reviewed suggests ...”
- “Based on these sources, the most likely interpretation is ...”
- “This appears stronger in X than in Y because ...”
### Bad phrasing
- “It is proven that ...” (when the evidence is actually mixed)
- “Experts agree ...” (without named support)
- “Clearly / obviously ...” (when it is not)
---
## Limitations & trade-offs
This section is mandatory.
Include:
- where the evidence is thin, dated, indirect, or conflicting
- what important conditions could flip the conclusion
- operational costs / risks / failure modes when relevant
- what the report did not establish
A report without limitations is not production-ready.
---
## Brief template
```markdown
# {Title}
## Answer
{Direct answer with citations.}
## Key evidence
- ... [1][2]
- ... [3]
## Limitations / trade-offs
- ... [4]
- ... [5]
## Non-obvious insight
{One useful nuance or challenged assumption.}
## References
[1] ...
```
---
## Full report template skeleton
```markdown
# {Title}
> Date: YYYY-MM-DD | Output: Brief / Full / Delta | Sources: N
> Stakes: Low / Medium / High | Evaluation: Passed / Not run / Needs review
## Executive Summary
- Direct answer
- Why it matters
- Main trade-off
- Confidence / boundary
## Main Analysis
...
## Limitations & Trade-offs
...
## Decision Framework
... (only if needed)
## References
...
```
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |