multi-agent-supervisor
Skill0 starsChanged 6 months agoMITView on GitHub
Implements the supervisor pattern: one orchestrator routes tasks to specialist agents (research, code, data). Use for multi-agent systems, delegating subtasks, LangGraph supervisor, or MCP-backed specialist workers.
In this repository14 other files it gives its agents.
- a2a-agent-delegationSkill
- agent-memory-patternsSkill
- agent-observabilitySkill
- agent-redteam-securitySkill
- browser-agent-automationSkill
- build-mcp-serverSkill
- create-production-skillSkill
- cursor-sdk-automationSkill
- deploy-agent-productionSkill
- evaluate-agent-qualitySkill
- github-ci-agentsSkill
- langgraph-multi-agentSkill
- rag-agent-pipelineSkill
- secure-mcp-serverSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 35
Stop when user goal satisfied.
- Distinctive: line 55
No change to other specialists
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Multi-Agent Supervisor
Architecture
User → Supervisor (routing only) → Specialist A | B | C → Supervisor → FINISHRule: Supervisor never executes tools directly — only picks the next worker.
Specialist design
| Specialist | Tools (via MCP) | Output |
| research | web_search, fetch_page | citations + summary |
| code | run_tests, lint, edit_files | patch or PR-ready diff |
| data | sql_query (read-only) | tables + interpretation |
Each specialist is a small ReAct or tool-calling subgraph with 2–5 tools max.
Routing prompt (supervisor)
Include in system message:
- List of workers with one-line capability
- Output format: exactly one of
[research, code, data, FINISH] - Stop when user goal satisfied; do not loop specialists unnecessarily
LangGraph sketch
- Node
supervisor— LLM returns next worker name - Nodes
research,code,data— run specialist; return to supervisor - Conditional edge from supervisor to worker or END
Failure handling
- Unknown route → default to clarifying question node
- Specialist error → append error to state; supervisor may retry once or escalate
- Max iterations (e.g. 15) to prevent infinite delegation
Scale-out
Add specialists by:
- New MCP server with focused tools
- New graph node + supervisor prompt line
- No change to other specialists
Pair with a2a-agent-delegation for cross-runtime agents.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.