Circus MCP
Enables AI agents to manage processes via structured MCP tools, reducing token usage by 75-80% compared to shell commands.
README
Circus MCP
Cut 75-80% of AI agent debugging tokens in development cycle. Process management via MCP — structured tools replace shell commands.
75-80% Token Reduction for AI Agent Debugging in Development Cycle
AI agents debugging processes via raw shell commands (supervisorctl, ps, journalctl) burn most of their tokens on unstructured output parsing, repeated commands, and inter-step reasoning. Circus MCP replaces this with structured, bounded MCP tool responses.
| Raw Commands | Circus MCP | Reduction | |
|---|---|---|---|
| Tool calls per investigation | 8-12 | 3-5 | 60-70% |
| Tokens per investigation | 2,900-9,400 | 935-1,535 | ~75% |
| With retries (typical) | ~10,000+ | ~2,000 | ~80% |
| Retry cost scaling | Exponential | Linear | — |
Process Management via MCP
Circus MCP exposes process lifecycle operations as MCP tools. AI agents call structured tools instead of parsing shell output.
| Tool | Parameters | Description |
|---|---|---|
list_processes |
— | List all managed processes |
get_process_status |
name |
Process state and PID |
start_process |
name |
Start a process |
stop_process |
name |
Stop a process |
restart_process |
name |
Restart a process |
add_process |
name, command, numprocesses?, working_dir? |
Add a new process dynamically |
Claude Code
claude mcp add circus-mcp -- uv run circus-mcp mcp
VS Code / Cursor
.vscode/mcp.json:
{
"servers": {
"circus-mcp": {
"command": "uv",
"args": ["run", "circus-mcp", "mcp"]
}
}
}
Circus MCP vs Supervisord MCP
| Circus MCP | Supervisord MCP | |
|---|---|---|
| Dynamic process addition | Via API | Not supported (requires config file edit + reload) |
| Log retrieval | stdout + stderr in one call | Separate calls |
| System stats (CPU/memory) | Available | Not available |
| Idempotent operations | ensure_started / ensure_stopped |
Throws error if already running |
| Transport | ZeroMQ (async) | HTTP XML-RPC (sync) |
| Best for | AI agent workflows | Existing Supervisord environments |
Documentation
- AI Token Reduction Solution — Token cost analysis, team-scale projections, research references
- AI & MCP Technical Background — Architecture, MCP hosting, tenant isolation
- Installation & CLI Reference — Setup, configuration, full command reference
License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.