cMCP

cMCP sits between agents and their tools so calls can be reviewed and denied, while generating a signed receipt of the decision. This gives teams an auditable trail for agent actions — useful for compliance, debugging, and guardrailing autonomous systems.

AI ToolFree Tier: Free (open source)Company: CommunityCategory: Safety & EthicsOpen Source: YesQuick Start: Install cMCP and wire it into your MCP server configuration to start logging denials.

cMCP

MCP-layer control tool that can deny an AI agent's tool call and emit a signed, tamper-evident receipt.

Visit cMCP →

cMCP sits between agents and their tools so calls can be reviewed and denied, while generating a signed receipt of the decision. This gives teams an auditable trail for agent actions — useful for compliance, debugging, and guardrailing autonomous systems.

FREE TIER
Free (open source)
COMPANY
Community
CATEGORY
Safety & Ethics › Agent Governance
OPEN SOURCE
Yes
PRIVACY
Local-first; receipts are verifiable artifacts
TAGS
mcpauditagent-safety

QUICK START

Install cMCP and wire it into your MCP server configuration to start logging denials.

Finding similar tools…
← Back to all tools