AgentGuard

Security & Safety 💻 Python ⚖️ Apache-2.0 🔴 No recent commits
99 stars

Agent Guard is a security toolset from CyberArk for AI agents. It provides secure secrets retrieval from external providers like AWS Secrets Manager and CyberArk Secrets Manager, and enables traceability of AI agent MCP communications via its MCP Proxy. It is intended for developers and security teams building or managing AI agents.

✨ Key features

  • Secure secrets retrieval from AWS Secrets Manager and CyberArk Secrets Manager.
  • MCP Proxy for traceability of AI agent MCP communications.
  • Available as Python libraries, CLI, and containerized solution.
  • Open-source project with contribution guidelines.
  • Contact support via GitHub issues.

🎯 Use cases

  • Securely fetch secrets for AI agents from external secret providers.
  • Monitor and trace MCP communications between AI agents and tools.
  • Integrate secret management into agent workflows using Python libraries.
  • Deploy Agent Guard as a container for scalable security enforcement.

❓ FAQ

What secret providers does Agent Guard support?

Agent Guard supports AWS Secrets Manager and CyberArk Secrets Manager.

How can I use Agent Guard?

You can use it as Python libraries, a CLI, or a containerized solution, as described in the linked documentation.

How can I contribute to Agent Guard?

Read the CONTRIBUTING.md file in the repository for contribution guidelines.

How can I contact the developers?

You can contact them via GitHub issues.

📊 Repository

Stars★ 99
Forks🍴 7
Open issues🐛 1
Last commit🕒 Sep 15, 2025
Created📅 Mar 2025
Language💻 Python
License⚖️ Apache-2.0

🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.