agent-overview — Claude Code subagent: what it does, how to install
Multi-agent architecture mapping nine specialized research roles in Claude Code. Coordinates literature reviews, GitHub analysis, and data synthesis.
npx claude-code-templates@latest --agent deep-research-team/agent-overviewWhat agent-overview does
This blueprint defines the Open Deep Research Team workflow. It organizes nine dedicated roles under a central orchestrator. The lead agent routes tasks, enforces quality checkpoints, and logs milestones with TodoWrite. Before work starts, a clarifier scores query ambiguity on a 0.0 to 1.0 scale and asks multiple-choice follow-ups.
Downstream agents handle distinct source types. The academic researcher extracts papers from ArXiv, PubMed, and Google Scholar while tracking DOIs. Meanwhile, the technical researcher inspects GitHub repositories and documentation. Without configuring the companion subagents, this file serves primarily as an operational blueprint.
What it helps with
- structure a multi-agent research workflow
- clarify ambiguous user queries with confidence scoring
- gather academic citations across ArXiv and PubMed
- coordinate technical reviews of GitHub repositories
Example prompts for the agent
- Show the multi-agent workflow for researching distributed consensus algorithms
- Use the agent-overview blueprint to plan an analysis of our repository stack
How to install agent-overview
1claude-code-templates
npx claude-code-templates@latest --agent deep-research-team/agent-overview2Manually: put the files in place
.claude/agents/agent-overview.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.codex/agents/agent-overview.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.cursor/agents/agent-overview.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about agent-overview
How to install agent-overview in Claude Code?
Run npx claude-code-templates@latest --agent deep-research-team/agent-overview in your terminal. You can also place the configuration directly at .claude/agents/agent-overview.md.
What does agent-overview do?
It outlines the orchestration and roles for a nine-agent research pipeline. The layout coordinates clarifying queries, gathering scholarly and code sources, and synthesizing findings.
Which data sources does this team structure cover?
The structure assigns academic searches to ArXiv, PubMed, and Google Scholar, while technical investigations target GitHub repositories and API documentation.