Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Performs comprehensive, multi-layered research on any topic with structured analysis and synthesis of information from multiple sources. Use when the user needs thorough investigation, market research, technical deep-dives, due diligence, or comprehensive analysis on any subject.
.claude/skills/bilal140202-deep-researcher/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-16 | ✗→✓ | ▲ Improved | 20% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 26% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 20% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 68% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 20% | 0% |
A comprehensive research workflow skill that guides Claude through systematic, multi-layered research on any topic.
Use this skill when the user:
Start by clearly establishing research parameters:
Document these parameters before proceeding.
Begin with broad reconnaissance to map the landscape:
Create an initial research map with main themes and subtopics.
For each identified key area, conduct detailed investigation:
Document findings systematically, noting sources and credibility.
Ensure accuracy and reliability:
Transform collected data into insights:
Create comprehensive, well-organized final report:
## Report Structure
1. **Executive Summary**
- Brief overview of research question
- Key findings (3-5 main points)
- Main conclusions
2. **Introduction**
- Research scope and objectives
- Methodology overview
- Key areas explored
3. **Main Body** (organized by themes)
- Detailed findings for each key area
- Supporting evidence and examples
- Multiple perspectives where relevant
- Data and statistics
4. **Analysis & Synthesis**
- Patterns and connections identified
- Cross-cutting themes
- Insights and interpretations
5. **Conclusions**
- Direct answers to research questions
- Summary of most significant findings
- Limitations of the research
6. **Sources & References**
- List of key sources consulted
- Credibility notes where relevantUse these frameworks as appropriate:
Copy and update this checklist as you work:
Research Progress:
- [ ] Step 1: Define scope and objectives
- [ ] Step 2: Initial exploratory research
- [ ] Step 3: Deep dive into key areas
- [ ] Step 4: Cross-reference and validate
- [ ] Step 5: Synthesize and analyze
- [ ] Step 6: Generate structured report| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-17 | fail→fail | 33,980 | 33,999 | +0% | 1 | 1 | 0% | 6,186 | 7,408 | +20% | 0 | 0 | — |
case-07 | fail→fail | 39,912 | 39,913 | +0% | 1 | 1 | 0% | 6,207 | 7,429 | +20% | 0 | 0 | — |
case-08 | fail→fail | 40,669 | 99,505 | +145% | 1 | 1 | 0% | 6,195 | 7,417 | +20% | 0 | 0 | — |
case-09 | fail→fail | 38,537 | 36,887 | -4% | 1 | 1 | 0% | 6,182 | 6,880 | +11% | 0 | 0 | — |
case-16 | fail→pass | 39,574 | 38,252 | -3% | 1 | 1 | 0% | 6,177 | 7,400 | +20% | 0 | 0 | — |
case-06 | fail→fail | 38,440 | 40,511 | +5% | 1 | 1 | 0% | 6,225 | 7,447 | +20% | 0 | 0 | — |
case-01 | pass→pass | 4,316 | 4,297 | -0% | 1 | 1 | 0% | 646 | 1,836 | +184% | 0 | 0 | — |
case-02 | pass→pass | 3,589 | 3,848 | +7% | 1 | 1 | 0% | 730 | 1,982 | +172% | 0 | 0 | — |
case-03 | pass→pass | 3,512 | 2,819 | -20% | 1 | 1 | 0% | 590 | 1,723 | +192% | 0 | 0 | — |
case-04 | fail→fail | 36,827 | 38,986 | +6% | 1 | 1 | 0% | 6,229 | 7,450 | +20% | 0 | 0 | — |
case-05 | fail→fail | 39,332 | 40,075 | +2% | 1 | 1 | 0% | 6,221 | 7,443 | +20% | 0 | 0 | — |
case-10 | pass→fail | 39,582 | 39,697 | +0% | 1 | 1 | 0% | 6,196 | 7,418 | +20% | 0 | 0 | — |
case-11 | fail→fail | 37,111 | 39,404 | +6% | 1 | 1 | 0% | 5,847 | 7,403 | +27% | 0 | 0 | — |
case-12 | fail→fail | 36,964 | 40,091 | +8% | 1 | 1 | 0% | 6,192 | 7,414 | +20% | 0 | 0 | — |
case-13 | pass→pass | 37,251 | 38,057 | +2% | 1 | 1 | 0% | 6,175 | 7,397 | +20% | 0 | 0 | — |
case-14 | fail→fail | 39,913 | 37,886 | -5% | 1 | 1 | 0% | 6,177 | 6,905 | +12% | 0 | 0 | — |
case-15 | fail→pass | 31,081 | 32,763 | +5% | 1 | 1 | 0% | 4,936 | 6,215 | +26% | 0 | 0 | — |
case-18 | fail→fail | 36,807 | 35,152 | -4% | 1 | 1 | 0% | 6,175 | 7,397 | +20% | 0 | 0 | — |
case-19 | fail→pass | 35,689 | 33,860 | -5% | 1 | 1 | 0% | 6,178 | 7,400 | +20% | 0 | 0 | — |
case-20 | fail→pass | 13,294 | 13,357 | +0% | 1 | 1 | 0% | 2,055 | 3,444 | +68% | 0 | 0 | — |
case-21 | fail→pass | 37,548 | 39,833 | +6% | 1 | 1 | 0% | 6,179 | 7,401 | +20% | 0 | 0 | — |
case-22 | fail→fail | 34,541 | 35,223 | +2% | 1 | 1 | 0% | 5,801 | 7,397 | +28% | 0 | 0 | — |
case-23 | fail→fail | 32,545 | 34,033 | +5% | 1 | 1 | 0% | 4,750 | 6,292 | +32% | 0 | 0 | — |
case-24 | fail→pass | 38,567 | 37,206 | -4% | 1 | 1 | 0% | 6,187 | 7,409 | +20% | 0 | 0 | — |
case-25 | fail→fail | 35,642 | 36,528 | +2% | 1 | 1 | 0% | 6,182 | 7,404 | +20% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 25 cases were attempted. The headline lift of +20 percentage points is the difference between those two pass rates over the 25 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.