Skip to content
Development
Command

/research

Called by team.md router when action is `research`. L1-heavy action: most agents are in the research layer.

From plugin
coco
26441 skills37 agents41 commands
Install
$ npx -y skills add coco-research/coco --agent claude-code

How 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
Ships withcoco

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.

Get the whole plugin