multi-agent-architect - Design and update multi-agent systems
Designs, debugs, optimizes, and updates multi-agent systems with LangGraph, LangChain, and DeepAgents while preserving copied files and provenance.
Tags
Updated: 2026-10-07Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Clarify workflow goals
- Define shared state schemas
- Design agent roles
- Build LangGraph routing
- Integrate memory and tools
- Debug existing agent systems
- Validate workflow changes
- Preserve provenance records
- Prepare handoff decisions
Inputs
- User goals
- Workflow scope
- Repository context
- Copied upstream files
- Provenance files
- Existing agent systems
Outputs
- Agent workflow designs
- Updated agent code
- Validation findings
- Provenance notes
- Handoff recommendations
- Reviewer packets
Requirements
- LangGraph, LangChain, or DeepAgents support