modus-agendi/managed-agent-control-mcp
by Various
Start, observe, and interact with Claude Managed Agents from any MCP client (Claude.ai, Claude Code, …). Pluggable auth (bearer/OIDC/Cognito); runs on stdio, Docker, or AWS Lambda.
MCP
modus-agendi/managed-agent-control-mcp
Added 7 June 2026
Overview
This tool provides an MCP server to start, observe, and interact with Claude Managed Agents from any MCP client such as Claude.ai or Claude Code. It supports pluggable authentication via bearer token, OIDC, or Cognito and can run on stdio, Docker, or AWS Lambda.
Best for
Best for
Developers who need to automate and observe Claude Managed Agents from MCP-compatible environments
Use cases
- Programmatically launch and monitor Claude Managed Agents from custom MCP clients
- Integrate agent lifecycle control into existing CI/CD or automation pipelines
- Deploy agent management endpoints on serverless infrastructure using AWS Lambda
How to use
Install
uv run python -m managed_agents_mcp Tools exposed
ANTHROPIC_API_KEYANTHROPIC_BASE_URLMCP_AUTH_MODEMCP_BEARER_TOKENMCP_OIDC_ISSUERMCP_OIDC_JWKS_URLMCP_OIDC_AUDIENCEMCP_ALLOWLIST_AGENTS_ACTIVEMCP_ALLOWED_AGENT_IDSMCP_ALLOW_DESTRUCTIVE
Tested with
Claude.ai, Claude Code, Cursor, mcp-remote, your own agent
Notes
This tool provides an MCP server to start, observe, and interact with Claude Managed Agents from any MCP client such as Claude.ai or Claude Code. It supports pluggable authentication via bearer token, OIDC, or Cognito and can run on stdio, Docker, or AWS Lambda.
0 stars on GitHub. Last updated 2026-06-03. Licensed MIT.
Use cases
- Programmatically launch and monitor Claude Managed Agents from custom MCP clients
- Integrate agent lifecycle control into existing CI/CD or automation pipelines
- Deploy agent management endpoints on serverless infrastructure using AWS Lambda
Pros
- Pluggable authentication supports bearer, OIDC, and Cognito for flexible security
- Runs on multiple deployment targets (stdio, Docker, Lambda) to match infrastructure
- Works with any MCP client, including Claude.ai and Claude Code
Cons
- Limited to Claude Managed Agents, not a general-purpose agent control tool
- Relatively new with no stars, indicating low community adoption or maturity
- Requires MCP client setup and understanding of the MCP protocol
Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.
Pros
- Pluggable authentication supports bearer, OIDC, and Cognito for flexible security
- Runs on multiple deployment targets (stdio, Docker, Lambda) to match infrastructure
- Works with any MCP client, including Claude.ai and Claude Code
Cons
- Limited to Claude Managed Agents, not a general-purpose agent control tool
- Relatively new with no stars, indicating low community adoption or maturity
- Requires MCP client setup and understanding of the MCP protocol
Pairs with
Other entries in the index that connect to this one. Click through to see the chain.
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.
