MCP server for AgentShield that classifies untrusted text for prompt injection, jailbreak, and social-engineering attempts before an agent processes it. It exposes a single tool, `classify_text`, which sends text to the hosted AgentShield API and returns a verdict with confidence and category.
From the registry: Detect prompt injection, jailbreak, and social-engineering attacks in LLM agents.
Please install the `agentshield-mcp` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Install Node.js and npm/npx** — The server is launched via npx, so Node.js with npm/npx must be installed on the system. (https://nodejs.org/)
- **Get an AgentShield API key** — Sign up for AgentShield and obtain an API key for the hosted classifier service. (https://agentshield.pro/signup)
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["-y","@eigenart/agentshield-mcp"]
- required environment variables:
- `AGENTSHIELD_API_KEY`: Your API key from agentshield.pro. (example: `<your-api-key>`)
- optional environment variables:
- `AGENTSHIELD_BASE_URL`: Override the default AgentShield API base URL, for example when using a self-hosted gateway. (example: `<your-base-url>`)
Note: README states the same command/args/env block works across MCP clients. Default service endpoint is the hosted AgentShield API; AGENTSHIELD_BASE_URL is only needed to override it.
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.AGENTSHIELD_API_KEYrequiredAPI key for AgentShield.Real-time infrastructure monitoring with metrics, logs, alerts, and ML-based anomaly detection.
Trending hip-hop artist momentum scores across four cultural dimensions.
Web scraping with stealth HTTP, real browsers, and Cloudflare bypass. CSS selectors supported.