What it does
The publisher describes it as a way to require human approval for AI agents. It also stores secrets with end-to-end encryption, and it gates access to Gmail, GitHub and MCP tools.
It suits setups where an agent's use of email, code hosting or other MCP tools should pass a person's check first. The available description does not cover installation steps, supported clients or how approval requests are handled. The project is hosted on GitHub.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 18 Nov 2026 · its page on the source
aiSecurityWarden
Human approval for AI agents: E2E-encrypted secrets and gated Gmail, GitHub and MCP tools.
v0.7.12As shown on registry.modelcontextprotocol.io when we last read it