DJD Trust MCP Server

Agent trust scoring, certification, and governance tools for the agent economy. Powered by DJD AgentScore.

Published package: djd-mcp-trust

Quick Start

Print a ready-to-paste MCP config snippet:

npx -y djd-mcp-trust install

Claude Code / Claude Desktop

Add to your MCP config (~/.claude/claude_desktop_config.json or Claude Code settings):

{
  "mcpServers": {
    "djd-trust": {
      "command": "npx",
      "args": ["-y", "djd-mcp-trust"],
      "env": {
        "DJD_API_KEY": "your-api-key-here"
      }
    }
  }
}

DJD_API_KEY is optional. Free tools work without it.

Local Development

{
  "mcpServers": {
    "djd-trust": {
      "command": "node",
      "args": ["/absolute/path/to/djd-mcp-server/dist/index.js"],
      "env": {
        "DJD_API_KEY": "your-api-key-here"
      }
    }
  }
}

Tools

| Tool | Description | API Key Required | |------|-------------|:---:| | check_trust | Check an agent's trust score before transacting | No | | check_certified | Full assessment with dimensions, sybil detection, trajectory | Yes (falls back to basic) | | get_trusted_providers | Find trusted agents above a score threshold | No | | report_bad_actor | Report a fraudulent or underperforming agent | Yes ($0.02) | | get_agent_info | Get a registered agent's full profile + fraud check | Partial | | register_agent | Register an agent in the DJD trust network | No |

check_trust

Check an agent or endpoint's trust score before transacting.

Input: { wallet: "0x..." } Returns: Score (0-100), tier, confidence, recommendation

check_certified

Full agent assessment with score dimensions, sybil detection, gaming indicators, and trajectory. Shows eligibility for DJD Certify tiers (coming Q2-Q3 2026).

Input: { wallet: "0x..." } Returns: Full score breakdown or basic score with certification eligibility info

get_trusted_providers

Find DJD-trusted providers above a minimum score threshold.

Input: { capability?: "image-generation", min_score?: 70 } Returns: Ranked list of registered agents with scores and verification badges

report_bad_actor

Report a fraudulent or underperforming agent. Affects the target's trust score.

Input: { wallet: "0x...", reason: "failed_delivery", details: "Description of issue" } Reasons: failed_delivery, payment_fraud, impersonation, malicious_behavior, other

get_agent_info

Get a registered agent's profile and fraud history.

Input: { wallet: "0x..." } Returns: Name, description, GitHub, verification status, fraud reports (if API key set)

register_agent

Register an agent in the DJD trust network for scoring and certification eligibility.

Input: { wallet: "0x...", name?: "My Agent", description?: "What it does", github_url?: "https://github.com/...", website_url?: "https://..." }

Environment Variables

| Variable | Required | Default | Description | |----------|:--------:|---------|-------------| | DJD_API_URL | No | https://djd-agent-score.fly.dev | AgentScore API base URL | | DJD_API_KEY | No | — | API key for paid endpoints. Free tools work without it. |

Development

npm install
npm run build
npm run dev  # runs with tsx for development
npx -y djd-mcp-trust --help

About DJD

DJD Services LLC builds governance infrastructure for the agent economy. AgentScore is the trust and reputation layer — the credit bureau for AI agents.

• AgentScore API: https://djdagentscore.dev
• DJD Governance Stack: Certification, forensics, security, and compliance for agent deployments