/research
Called by team.md router when action is `research`. L1-heavy action: most agents are in the research layer.
$ npx -y skills add coco-research/coco --agent claude-codeHow it fires
How this command gets triggered: by you, by Claude, or both.
- Fires itselfClaude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/research
Context preview
What this command does when you run it.
Called by team.md router when action is `research`. L1-heavy action: most agents are in the research layer.
Command definition
research.md/team research — Research Pipeline
> Called by team.md router when action is `research`. > L1-heavy action: most agents are in the research layer.
Role Selection Bias
| Layer | Preferred Roles | Count | |-------|----------------|-------| | L1 | domain-researcher, technical-analyst, business-analyst, security-analyst, ux-researcher | 4-6 | | L2 | technical-writer (compiles findings) | 1-2 | | L3 | domain-accuracy, doc-quality | 1 | | L4 | principal-architect or principal-pm (topic-dependent) | 1 |
Research Angle Assignment
Each L1 agent gets a specific research angle to prevent overlap:
- **domain-researcher** → Industry context, competitive landscape, prior art
- **technical-analyst** → Technical feasibility, architecture options, dependency analysis
- **business-analyst** → Business case, stakeholder impact, requirements implications
- **security-analyst** → Security implications, compliance requirements, threat landscape
- **ux-researcher** → User impact, usability considerations, accessibility implications
Pipeline Customization
Layer 1: Deep Research (Primary Layer)
- **Mode:** `default` (read-only + web tools)
- Each agent uses WebSearch, WebFetch, and project file reads
- Agents produce structured findings with sources and confidence levels
- This is the main work layer — expect 70% of effort here
Layer 2: Compilation
- **Mode:** `default`
- technical-writer compiles all L1 findings into a unified research document
- Structure: Summary → Analysis by Angle → Cross-Cutting Themes → Recommendation
- Deduplicates and resolves conflicting findings
**Toolkit integration:**
- Check team:toolkit.md for document compilation tools if producing a formal research report
- If outputting to Confluence, check "Confluence Publishing" entry for API notes
Layer 3: Accuracy Check
L3 domain-accuracy verifies:
- Source URLs are valid and claims match sources
- Technical claims are accurate
- No outdated information presented as current
Layer 4: Strategic Synthesis
Principal produces:
- Executive summary for decision-makers
- Ranked recommendations with trade-offs
- Identified unknowns and suggested next steps
GSD Integration
N/A — Research is project-context-independent.
Read more
/team research — Research Pipeline
> Called by team.md router when action is `research`. > L1-heavy action: most agents are in the research layer.
Role Selection Bias
| Layer | Preferred Roles | Count | |-------|----------------|-------| | L1 | domain-researcher, technical-analyst, business-analyst, security-analyst, ux-researcher | 4-6 | | L2 | technical-writer (compiles findings) | 1-2 | | L3 | domain-accuracy, doc-quality | 1 | | L4 | principal-architect or principal-pm (topic-dependent) | 1 |
Research Angle Assignment
Each L1 agent gets a specific research angle to prevent overlap:
- **domain-researcher** → Industry context, competitive landscape, prior art
- **technical-analyst** → Technical feasibility, architecture options, dependency analysis
- **business-analyst** → Business case, stakeholder impact, requirements implications
- **security-analyst** → Security implications, compliance requirements, threat landscape
- **ux-researcher** → User impact, usability considerations, accessibility implications
Pipeline Customization
Layer 1: Deep Research (Primary Layer)
- **Mode:** `default` (read-only + web tools)
- Each agent uses WebSearch, WebFetch, and project file reads
- Agents produce structured findings with sources and confidence levels
- This is the main work layer — expect 70% of effort here
Layer 2: Compilation
- **Mode:** `default`
- technical-writer compiles all L1 findings into a unified research document
- Structure: Summary → Analysis by Angle → Cross-Cutting Themes → Recommendation
- Deduplicates and resolves conflicting findings
**Toolkit integration:**
- Check team:toolkit.md for document compilation tools if producing a formal research report
- If outputting to Confluence, check "Confluence Publishing" entry for API notes
Layer 3: Accuracy Check
L3 domain-accuracy verifies:
- Source URLs are valid and claims match sources
- Technical claims are accurate
- No outdated information presented as current
Layer 4: Strategic Synthesis
Principal produces:
- Executive summary for decision-makers
- Ranked recommendations with trade-offs
- Identified unknowns and suggested next steps
GSD Integration
N/A — Research is project-context-independent.
Meet Coco. A superintelligent agent framework powered by an advisory board of 389 world-class minds. Scale your AI assistant into a complete engineering department with 142 skills, 277 commands, and persistent state. Universal compatibility. Local privacy. Free and open source.
Repo: coco-research/coco
Other commands on coco.
- /mermaid
Build beautiful Mermaid diagrams using beautiful-mermaid. Covers all 6 diagram types, theming, SVG/ASCII output, and CoCo Platform integration.
Open command - /read
Show latest emails from a specific person. Usage: /email-read alice
Open command - /reply
Draft a reply to a specific email. Usage: /email-reply Project Phase 2 Contract Data
Open command - /save
Save matching emails to a project folder for sync processing. Usage: /email-save alice to emails/
Open command - /search
Search emails by subject keywords. Usage: /email-search github webhook
Open command - /summary
AI summary of today's emails — key decisions, action items, meetings. No arguments needed.
Open command

