The agent-immune MCP server provides adaptive threat intelligence for AI agent security, focusing on semantic memory, multi-turn escalation, output scanning, rate limiting, and prompt hardening. It is designed to enhance existing governance stacks by learning from incidents and detecting rephrased attacks.
From the registry: AI agent security: prompt injection detection, semantic memory, output scanning, prompt hardening
$ pip install 'agent-immune[mcp]'https://pypi.org/project/agent-immune/Please install the `agent-immune` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install agent-immune with MCP extras** — Install via pip with MCP support Run: `pip install 'agent-immune[mcp]'` (https://pypi.org/project/agent-immune/)
Canonical MCP server config (stdio transport):
- command: `python`
- args: ["-m","agent_immune","serve","--transport","stdio"]
Note: Alternative HTTP transport: python -m agent_immune serve --transport http --port 8000. Tools: assess_input, assess_output, learn_threat, harden_prompt, get_metrics.
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.API_KEYrequiredAPI key for accessing external services.Dead code, security, secrets detection and code quality for Python, TypeScript, Go.