Add AgentShield — open-source AI agent firewall and audit log

[brigen]

Description

Adding AgentShield to the AI Agent section.

AgentShield is an open-source security layer for AI agents:

• Sits between agents and their tools as a policy-enforcing proxy
• Intercepts MCP tool calls, CLI commands, and HTTP requests
• Rules-based policy engine: allow/deny/warn with rate limiting and regex argument matching
• Structured JSON audit logging with automatic secret redaction (AWS keys, GitHub tokens, API creds)
• Real-time web dashboard for monitoring agent activity
• Blocks data exfiltration and destructive CLI operations (rm -rf, git push --force)

MIT licensed, installed via npm, actively maintained.

[github-actions]

Hi @brigen! 👋

Thank you for your pull request! We're excited to review your contribution.

Please make sure to follow our recommendation template when submitting a tool recommendation.

We'll review your PR as soon as possible! 🚀