HallucC MCP Endpoint ● online

Claim-by-claim AI hallucination detection with sources · agent trajectory verification · L0–L3 risk gating for Computer-Use actions · Agent source code static audit · prompt-injection / jailbreak guard.

This URL is a Model Context Protocol endpoint (Streamable HTTP). It is meant for MCP clients, not browsers — point your client at it:

https://aihcc.cloud/mcp

1. Get an API key

Sign up at aihcc.cloud and create a key on the Keys page. The free tier includes a daily quota.

2. Connect your client

Claude Code

claude mcp add --transport http hallucc https://aihcc.cloud/mcp \
  --header "Authorization: Bearer <your HallucC API key>"

Cursor / Claude Desktop (mcp.json)

{
  "mcpServers": {
    "hallucc": {
      "url": "https://aihcc.cloud/mcp",
      "headers": { "Authorization": "Bearer <your HallucC API key>" }
    }
  }
}

Or run locally via npx (stdio transport)

{
  "mcpServers": {
    "hallucc": {
      "command": "npx",
      "args": ["-y", "hallucc-mcp"],
      "env": { "HALLUCC_API_KEY": "<your HallucC API key>" }
    }
  }
}

Tools

• verify_text — claim-by-claim hallucination check with cited sources
• verify_agent — verify an agent's tool-call trajectory
• check_cua_actions — L0–L3 risk gating before Computer-Use actions execute
• check_cua_code_audit — static audit of Agent source code for dangerous patterns
• check_safety — 40+ feature prompt-injection / jailbreak screen

GitHub · Glama · Dashboard
Health check: GET /health · MCP clients should POST /mcp