Aegis is a local-first credential isolation proxy for AI agents that injects secrets at the network boundary, preventing agents from seeing or transmitting real credentials.
From the registry: Credential isolation for AI agents. Inject secrets at the network boundary.

$ npx -y @getaegis/cli inithttps://github.com/getaegis/aegisPlease install the `aegis-getaegis` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Node.js 20+** — Required for the @getaegis/cli package. (https://nodejs.org)
- **Initialize the Aegis vault** — Run `aegis init` once to create the encrypted vault and store the master key in the OS keychain. Run: `npx -y @getaegis/cli init` (https://github.com/getaegis/aegis)
- **Add credentials and agent grants** — Use `aegis vault add` and `aegis agent add/grant` to enroll services before the MCP server can inject them. (https://github.com/getaegis/aegis)
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["-y","@getaegis/cli","mcp","serve"]
Note: Aegis injects credentials at the network boundary so the agent never sees raw API keys. Exposes `aegis_proxy_request`, `aegis_list_services`, and `aegis_health` tools.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.SLACK_TOKENrequiredToken for authenticating with Slack APISmart MCP proxy with BM25 tool discovery, quarantine security, and ~99% token savings