Skip to main content
Glama

@caelaralabs/shield-mcp

A stdio Model Context Protocol server for Caelara Shield. It exposes four tools that call https://shield.caelaralabs.com/v1/.... Plain Node ESM, no runtime dependencies, Node 18 or newer.

Shield results are heuristic checks, not guarantees.

Tools

Tool

Route

Price per call

shield_scan_prompt

POST /v1/shield/prompt

0.015 USDC

shield_validate_output

POST /v1/shield/output

0.015 USDC

shield_proxy_fetch

GET /v1/proxy/_?url=...

0.025 USDC

shield_audit_token

POST /v1/shield/tx

0.050 USDC

Input schemas match https://shield.caelaralabs.com/mcp.json.

Related MCP server: mcpsentry

Configure

Set CAELARA_KEY to your Caelara API key. It is sent as the X-Caelara-Key header. The key is read from the environment only.

Example MCP client configuration:

{
  "mcpServers": {
    "shield": {
      "command": "npx",
      "args": ["-y", "@caelaralabs/shield-mcp"],
      "env": { "CAELARA_KEY": "your-key" }
    }
  }
}

Optional environment variables:

  • CAELARA_SHIELD_URL: override the base URL (default https://shield.caelaralabs.com).

  • CAELARA_SHIELD_TIMEOUT_MS: request timeout in milliseconds (default 3000).

Behaviour

  • Stateless. Nothing is written to disk and request or response bodies are never logged.

  • Every upstream request has a 3000 ms timeout and sends User-Agent: CaelaraLabs-Engine/1.0.

  • Failures are returned as tool results with isError: true and a JSON body (timeout, network_error, upstream_error, invalid_arguments). The server does not throw to the client.

  • Without a valid key, a 402 response is returned as payment_required, including the X-Payment-Required header value and the response body so you can see the x402 terms. Paying via x402 happens on the REST routes (see https://shield.caelaralabs.com/x402-manifest.json), not through this server.

Test

node --test test/

The tests start a local fake HTTP server and exercise initialize, tools/list and tools/call (success, 402, timeout).

License

MIT. Caelara Labs, ops@caelaralabs.com.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Adaptive security for AI agents: assess inputs for prompt injection, scan outputs for credential/PII leaks, teach new attack patterns to semantic memory, harden prompts, and monitor metrics. Runs locally via MCP stdio.
      120 PyPI
      Apache 2.0
    • A
      license
      Not graded
      quality
      D
      maintenance
      Security scanner and runtime proxy for MCP servers. Catches tool poisoning, prompt injection, and rug-pull attacks (silent tool description changes) before they reach your AI agent. Includes a static scanner and a runtime stdio proxy.
      57 npm
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables MCP servers and local AI agents to safely interact with host environments by blocking indirect prompt injections, directory traversal, and malicious shell commands.
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Provides an MCP stdio interface for AI agents to propose tool calls that are validated against task policy, held for human approval when required, executed in a sandboxed Linux environment, and recorded in an audit log.
      MIT