Developer Hub

Build with SENTINEL

Zero-knowledge behavioral attestation for AI agents. Prove compliance without revealing data. First proof in 5 minutes.

MCP Integration

MCP-native integration for Claude. REST API for OpenAI (GPT), Google (Gemini), xAI (Grok), OpenClaw, and any AI agent.

Add SENTINEL to your Claude Desktop or Claude Code MCP config:

// Add to your MCP server configuration
{
  "sentinel": {
    "command": "npx",
    "args": ["-y", "@anthropic-ai/sentinel-mcp"],
    "env": {
      "SENTINEL_API_KEY": "your-api-key"
    }
  }
}

This gives you access to 8 MCP tools: prove, verify, chain, aggregate, compute BID, sanitize input, and more. Say "prove my agent is compliant" in Claude and SENTINEL handles the rest.

Using GPT, Gemini, Grok, OpenClaw, or another platform? Use the REST API directly — same capabilities, same results:

# Generate a compliance proof via REST API (works from any platform)
curl -X POST https://api.starksentinel.com/v1/prove \
  -H "X-API-Key: $SENTINEL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "observations": [95, 98, 92, 97, 100, 94, 96, 99],
    "baseline_mean": 96,
    "ucl": 110,
    "lcl": 80
  }'

Full walkthrough: Quickstart Guide · Complete reference: API Docs

< 8ms
P99 proof generation
5.9KB
Proof size
128-bit
Security level
7
API endpoints
8
MCP tools
0
Data leaked