multi-agent-openclaw

Design and run structured multi-agent research workflows locally.
System: checking…
Memory System Architecture
Intelligent plan–task–memory orchestration, dynamic model selection
User Input Search Implemented Vector Embed Implemented Vector Store Reorder Implemented Planning Model Self-select Task System Task Type Code / General Simple Self-select Medium Self-select Complex Self-select General Self-select Code Model Select Response Memory System Persistent Final Output Context Update Loop Memory Capacity ∞ Unlimited Implemented Roadmap/next
Step 1 · Define Task
Research Essay Mode

Task Clarity Check

No clarification yet.
Step 2 · Configure Workflow

Mode

Available Agents

Drag agents into role cards
Chats

Role Assignment

Assign one or more agents to each responsibility
Assignment JSON (secondary)
Step 3 · Run & Review
Quality Insights
Filter: none
Suggested Actions
No action generated yet.