<p align="center"> <img src="assets/logo.png" alt="det-acp logo" width="280" /> </p>

Self-Evolving Deterministic Agent Control Protocol

A governance gateway for AI agents — making every action bounded, auditable, reversible, and explainable.

Works transparently with Cursor, Claude Code, Codex, and any MCP-compatible agent. Also supports shell command governance and a language-agnostic HTTP API.

🛡️ Governance in Action

<table> <tr> <td align="center" colspan="2"> <img src="assets/mcp-config.gif" alt="Setting up governance rule in Cursor" width="80%" /> <br /> <b>1. Set Up Governance Rule</b> <br /> _{Enable the governance rule in Cursor's settings to protect your workspace.} </td> </tr> <tr> <td align="center" width="50%"> <img src="assets/block-access-env.gif" alt="Blocking .env file access" width="90%" /> <br /> <b>2. Block Secrets Exfiltration</b> <br /> _{Agent attempts to read <code>.env</code> and write secrets — blocked instantly.} </td> <td align="center" width="50%"> <img src="assets/partial-block.gif" alt="Blocking credential search" width="90%" /> <br /> <b>3. Block Credential Scanning</b> <br /> _{Agent tries to search for credentials and secrets files — denied by policy.} </td> </tr> </table>