Agent Ledger API

EU AI Act compliance-as-a-service. Register your AI agents, classify risk automatically, and maintain an immutable audit trail of decisions — all in a few API calls.

What you can do

• Register agents — submit your AI agent's profile and receive an instant EU AI Act risk classification (unacceptable / high / limited / minimal) with relevant articles
• Log decisions — create tamper-evident audit records (SHA-256 hashed) of every automated decision your agent makes
• Run risk scans — interactive questionnaire-based scan to pre-check any AI system before deployment, no account required
• Manage API keys — create and revoke keys per project or environment

Try It Live (Sandbox)

Explore the API without creating an account. Use our demo agent and pre-authenticated sandbox key.

Pre-authenticated Swagger UI

Use the Swagger UI with sandbox credentials pre-filled:

Example Request to Sandbox Agent

curl -X GET https://agentledgerhq.com/api/v1/agents/agt_demo_sandbox_001 \
  -H "Authorization: Bearer al_sandbox_ed036aa88c32f53209cdc956b221f4cf" \
  -H "Content-Type: application/json"

const res = await fetch('https://agentledgerhq.com/api/v1/agents/agt_demo_sandbox_001', {
  headers: {
    'Authorization': 'Bearer al_sandbox_ed036aa88c32f53209cdc956b221f4cf',
    'Content-Type': 'application/json'
  }
})
const agent = await res.json()
console.log(agent)

Ready to build?

Create your own API key and agents:

Quick Start

From zero to your first compliant agent in under 2 minutes.

1. 1
   Create an account and get your API key

   cURL JavaScript

   curl -X POST https://agentledgerhq.com/api/v1/register \
     -H "Content-Type: application/json" \
     -d '{"name": "Ada Lovelace", "email": "ada@example.com"}'

   const res = await fetch('https://agentledgerhq.com/api/v1/register', {
     method: 'POST',
     headers: { 'Content-Type': 'application/json' },
     body: JSON.stringify({ name: 'Ada Lovelace', email: 'ada@example.com' })
   })
   const data = await res.json()
   // data.api_key = "al_live_xxx..." — store this securely!

   ⚠️

   Your API key is shown only once. Store it in a secrets manager or environment variable immediately. It will also be sent to your email.
2. 2
   Register your first AI agent

   cURL JavaScript

   curl -X POST https://agentledgerhq.com/api/v1/agents \
     -H "Authorization: Bearer al_live_YOUR_KEY" \
     -H "Content-Type: application/json" \
     -d '{
       "name": "Hiring Screening Bot",
       "description": "Screens job applications and shortlists candidates",
       "type": "decision",
       "autonomy_level": "semi-autonomous",
       "decision_types": ["application_screening", "candidate_ranking"],
       "data_processed": ["cv", "employment_history"],
       "human_oversight": true,
       "deployment_environment": "production"
     }'

   const res = await fetch('https://agentledgerhq.com/api/v1/agents', {
     method: 'POST',
     headers: {
       'Authorization': 'Bearer al_live_YOUR_KEY',
       'Content-Type': 'application/json'
     },
     body: JSON.stringify({
       name: 'Hiring Screening Bot',
       description: 'Screens job applications and shortlists candidates',
       type: 'decision',
       autonomy_level: 'semi-autonomous',
       decision_types: ['application_screening', 'candidate_ranking'],
       data_processed: ['cv', 'employment_history'],
       human_oversight: true,
       deployment_environment: 'production'
     })
   })
   const agent = await res.json()
   // agent.risk_class = "high", agent.eu_ai_act_articles = [...]
   console.log(`Agent ${agent.id} classified as: ${agent.risk_class}`)
   

3. 3
   Log a decision for your agent

   cURL JavaScript

   curl -X POST https://agentledgerhq.com/api/v1/decisions \
     -H "Authorization: Bearer al_live_YOUR_KEY" \
     -H "Content-Type: application/json" \
     -d '{
       "agent_id": "agent_xxx",
       "decision_type": "application_screening",
       "input_summary": "CV for John Doe, 5 years experience",
       "output_summary": "Shortlisted — meets minimum criteria",
       "confidence": 0.82,
       "human_reviewed": true
     }'

   const res = await fetch('https://agentledgerhq.com/api/v1/decisions', {
     method: 'POST',
     headers: {
       'Authorization': 'Bearer al_live_YOUR_KEY',
       'Content-Type': 'application/json'
     },
     body: JSON.stringify({
       agent_id: 'agent_xxx',
       decision_type: 'application_screening',
       input_summary: 'CV for John Doe, 5 years experience',
       output_summary: 'Shortlisted — meets minimum criteria',
       confidence: 0.82,
       human_reviewed: true
     })
   })
   const decision = await res.json()
   // decision.audit_hash = "sha256:abc..." — immutable proof of this decision

Try the API

Copy-paste these examples into your terminal to test the API. Replace al_live_YOUR_KEY with your actual API key.

Get Your API Key

curl -X POST https://agentledgerhq.com/api/v1/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Test Account",
    "email": "test@example.com"
  }'

# Response includes:
# "api_key": "al_live_XXXX..."

Check API Health

curl https://agentledgerhq.com/api/health

# Response: { "status": "ok", "service": "agentledger-api", "version": "1.0.0" }

Run a Risk Scan (No Auth Required)

curl -X POST https://agentledgerhq.com/api/v1/scans \
  -H "Content-Type: application/json" \
  -d '{
    "q1_purpose": "hiring_decisions",
    "q2_human_impact": "high",
    "q3_autonomy": "semi_autonomous",
    "q4_data_types": ["cv_data", "employment_history"],
    "q5_human_oversight": true,
    "q6_affected_persons": "job_applicants",
    "q7_eu_operated": true,
    "q8_reversible": true,
    "q9_sector": "HR"
  }'

# Response includes:
# "risk_class": "high"
# "risk_score": 72
# "applicable_articles": [...]

Register an Agent

curl -X POST https://agentledgerhq.com/api/v1/agents \
  -H "Authorization: Bearer al_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Credit Scoring Engine",
    "description": "Evaluates creditworthiness based on financial history",
    "type": "decision",
    "autonomy_level": "semi-autonomous"
  }'

# Response includes: "id", "risk_class", "risk_score", "eu_ai_act_articles"

List Your Agents

curl -H "Authorization: Bearer al_live_YOUR_KEY" \
  https://agentledgerhq.com/api/v1/agents

# Response: { "data": [...agents...], "total": 5 }

Log a Decision

curl -X POST https://agentledgerhq.com/api/v1/decisions \
  -H "Authorization: Bearer al_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "agent_id": "ag_XXXX",
    "decision_type": "credit_decision",
    "input_summary": "Applicant: Jane Doe, income €50k/year, credit score 750",
    "output_summary": "Approved for €20,000 loan",
    "confidence": 0.85,
    "human_reviewed": true
  }'

# Response includes: "id", "decision_risk_score", "audit_hash"

Export Decision Ledger

curl -H "Authorization: Bearer al_live_YOUR_KEY" \
  "https://agentledgerhq.com/api/v1/decisions/export?format=csv&agent_id=ag_XXXX"

# Returns CSV file for audit compliance

Authentication

Agent Ledger uses API key authentication via the HTTP Authorization header.

Bearer Token

Pass your API key in the Authorization header using the Bearer scheme:

Authorization: Bearer al_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Key Format

All API keys follow the pattern al_live_ followed by a 32-character hex string. Keys are prefixed to distinguish live keys from test keys in future.

Key Security

• Keys are shown only once at creation. Store them in a secrets manager (e.g. AWS Secrets Manager, HashiCorp Vault, or environment variables).
• Never commit keys to version control.
• Rotate keys by creating a new key and revoking the old one.
• Each account can have multiple keys — use one per environment or service.

Endpoints that require authentication

Rate Limits

Rate limits protect the API from abuse and ensure fair usage across all accounts.

When you exceed a rate limit, the API responds with HTTP 429 Too Many Requests. Check the Retry-After response header for the number of seconds to wait before retrying.

Errors

Agent Ledger uses conventional HTTP status codes. Error responses include a machine-readable error code and a human-readable message.

Error Response Format

{
  "error": "validation_error",
  "message": "name is required and must be a string (max 100 characters)"
}

HTTP Status Codes

EU AI Act Classification

The EU AI Act establishes a risk-based framework for AI systems operating in the European Union. Agent Ledger automatically classifies your agents into one of four tiers.

Applicable Articles

When you register an agent, Agent Ledger returns a list of applicable EU AI Act articles based on your agent's profile. Common articles include:

Risk Score Calculation

Agent Ledger calculates two types of risk scores: agent-level risk and decision-level risk. Both drive compliance recommendations.

Agent-Level Risk Score (0–99)

The agent risk score is calculated when you register a new agent. It assesses the inherent risk of your AI system across multiple dimensions:

Risk Class Mapping

Decision-Level Risk Score (0.0–1.0)

When you log decisions, each decision gets its own risk score independent of the agent's overall risk. This captures the actual risk of specific outputs in context:

Example Decision Risk

How This Impacts Your Compliance

• Agent risk > 70: Register your AI system in the EU AI Act database before deployment.
• Decision risk > 0.7: Flag for mandatory human review before implementation.
• Low confidence + no review: Treat as high-risk and document in your audit trail.
• Article 5 flagged: Stop deployment immediately — seek legal counsel.

Health Check

Verify that the API is operational.

curl https://agentledgerhq.com/api/health

const res = await fetch('https://agentledgerhq.com/api/health')
const data = await res.json()

{
  "status": "ok",
  "service": "agentledger-api",
  "version": "1.0.0",
  "timestamp": "2025-01-01T12:00:00.000Z"
}

API Info

Returns API version info and available endpoint groups.

curl https://agentledgerhq.com/api/v1

const res = await fetch('https://agentledgerhq.com/api/v1')
const info = await res.json()

Register Account

Create a new Agent Ledger account and receive your first API key.

curl -X POST https://agentledgerhq.com/api/v1/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Ada Lovelace",
    "email": "ada@example.com"
  }'

const res = await fetch('https://agentledgerhq.com/api/v1/register', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    name: 'Ada Lovelace',
    email: 'ada@example.com'
  })
})
const data = await res.json() // 201 Created

{
  "account_id": "acc_a1b2c3d4e5f6",
  "name": "Ada Lovelace",
  "email": "ada@example.com",
  "plan": "free",
  "api_key": "al_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx",
  "key_prefix": "al_live_xxxxxxxx...",
  "created_at": "2025-01-01T12:00:00.000Z",
  "message": "Account created. Your API key has been sent to your email address. Store it securely — it will not be shown again."
}

Risk Scan

Run a full EU AI Act risk assessment questionnaire. No account required — results are returned immediately.

curl -X POST https://agentledgerhq.com/api/v1/scans \
  -H "Content-Type: application/json" \
  -d '{
    "answers": {
      "q1_purpose": "hiring",
      "q2_human_impact": "high",
      "q3_autonomy": "semi_autonomous",
      "q4_data_types": ["financial"],
      "q5_human_oversight": true,
      "q6_affected_persons": "job_applicants",
      "q7_eu_operated": true,
      "q8_reversible": true,
      "q9_sector": "hr",
      "q10_existing_registration": false
    }
  }'

const res = await fetch('https://agentledgerhq.com/api/v1/scans', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    answers: {
      q1_purpose: 'hiring',
      q2_human_impact: 'high',
      q3_autonomy: 'semi_autonomous',
      q4_data_types: ['financial'],
      q5_human_oversight: true,
      q6_affected_persons: 'job_applicants',
      q7_eu_operated: true,
      q8_reversible: true,
      q9_sector: 'hr',
      q10_existing_registration: false
    }
  })
})
const scan = await res.json()

{
  "scan_id": "scan_a1b2c3d4e5f6",
  "risk_class": "high",
  "risk_score": 74,
  "eu_ai_act_classification": "Annex III — High-Risk AI System",
  "applicable_articles": [
    "Article 9 — Risk Management System",
    "Article 10 — Data and Data Governance",
    "Article 12 — Record-Keeping",
    "Article 14 — Human Oversight",
    "Annex III — High-Risk AI System"
  ],
  "action_plan": [
    {
      "priority": 1,
      "action": "Implement a risk management system covering the full AI lifecycle",
      "deadline": "Before deployment",
      "article": "Article 9"
    },
    {
      "priority": 2,
      "action": "Document training data sources, selection criteria, and quality measures",
      "deadline": "Before deployment",
      "article": "Article 10"
    },
    {
      "priority": 3,
      "action": "Set up automated logging of all system decisions",
      "deadline": "Before deployment",
      "article": "Article 12"
    }
  ],
  "report_available": false,
  "report_url": null,
  "upgrade_url": "https://agentledgerhq.com/#pricing",
  "scanned_at": "2025-01-01T12:00:00.000Z"
}

List Agents

Retrieve all AI agents registered under your account.

curl https://agentledgerhq.com/api/v1/agents \
  -H "Authorization: Bearer al_live_YOUR_KEY"

const res = await fetch('https://agentledgerhq.com/api/v1/agents', {
  headers: { 'Authorization': 'Bearer al_live_YOUR_KEY' }
})
const { agents } = await res.json()

{
  "agents": [
    {
      "id": "agent_a1b2c3d4e5f6",
      "name": "Hiring Screening Bot",
      "type": "decision",
      "risk_class": "high",
      "risk_score": 78,
      "compliance_status": "compliant",
      "deployment_environment": "production",
      "created_at": "2025-01-01T12:00:00.000Z"
    }
  ]
}

Get Agent

Retrieve a single agent by ID, including full profile and applicable EU AI Act articles.

curl https://agentledgerhq.com/api/v1/agents/agent_a1b2c3d4e5f6 \
  -H "Authorization: Bearer al_live_YOUR_KEY"

const res = await fetch('https://agentledgerhq.com/api/v1/agents/agent_a1b2c3d4e5f6', {
  headers: { 'Authorization': 'Bearer al_live_YOUR_KEY' }
})
const agent = await res.json()

{
  "id": "agent_a1b2c3d4e5f6",
  "name": "Hiring Screening Bot",
  "description": "Screens job applications and shortlists candidates",
  "type": "decision",
  "autonomy_level": "semi-autonomous",
  "risk_class": "high",
  "risk_score": 78,
  "eu_ai_act_articles": [
    "Article 9 — Risk Management System",
    "Article 10 — Data and Data Governance",
    "Article 12 — Record-Keeping",
    "Article 14 — Human Oversight",
    "Annex III — High-Risk AI System"
  ],
  "compliance_status": "compliant",
  "deployment_environment": "production",
  "human_oversight": true,
  "created_at": "2025-01-01T12:00:00.000Z"
}

Register Agent

Register a new AI agent and receive an instant EU AI Act risk classification.

curl -X POST https://agentledgerhq.com/api/v1/agents \
  -H "Authorization: Bearer al_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Loan Approval Engine",
    "description": "Automated credit scoring and loan approval for retail customers",
    "type": "decision",
    "autonomy_level": "semi-autonomous",
    "decision_types": ["credit_scoring", "loan_approval"],
    "data_processed": ["financial", "credit_history", "employment"],
    "human_oversight": true,
    "deployment_environment": "production",
    "metadata": {
      "model_version": "2.1.0",
      "team": "credit-risk"
    }
  }'

const res = await fetch('https://agentledgerhq.com/api/v1/agents', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer al_live_YOUR_KEY',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    name: 'Loan Approval Engine',
    description: 'Automated credit scoring and loan approval for retail customers',
    type: 'decision',
    autonomy_level: 'semi-autonomous',
    decision_types: ['credit_scoring', 'loan_approval'],
    data_processed: ['financial', 'credit_history', 'employment'],
    human_oversight: true,
    deployment_environment: 'production',
    metadata: { model_version: '2.1.0', team: 'credit-risk' }
  })
})
const agent = await res.json() // 201 Created

{
  "id": "agent_a1b2c3d4e5f6",
  "name": "Loan Approval Engine",
  "risk_class": "high",
  "risk_score": 82,
  "eu_ai_act_articles": [
    "Article 9 — Risk Management System",
    "Article 10 — Data and Data Governance",
    "Article 12 — Record-Keeping",
    "Article 14 — Human Oversight",
    "Annex III — High-Risk AI System"
  ],
  "status": "registered",
  "compliance_status": "compliant",
  "created_at": "2025-01-01T12:00:00.000Z"
}

Delete Agent

Permanently remove an agent and its classification data from your account.

curl -X DELETE https://agentledgerhq.com/api/v1/agents/agent_a1b2c3d4e5f6 \
  -H "Authorization: Bearer al_live_YOUR_KEY"

const res = await fetch('https://agentledgerhq.com/api/v1/agents/agent_a1b2c3d4e5f6', {
  method: 'DELETE',
  headers: { 'Authorization': 'Bearer al_live_YOUR_KEY' }
})
// 204 No Content on success

List Decisions

Retrieve the audit log of decisions made by your agents.

curl "https://agentledgerhq.com/api/v1/decisions?agent_id=agent_xxx&limit=20&offset=0" \
  -H "Authorization: Bearer al_live_YOUR_KEY"

const params = new URLSearchParams({
  agent_id: 'agent_xxx',
  limit: '20',
  offset: '0'
})
const res = await fetch(`https://agentledgerhq.com/api/v1/decisions?${params}`, {
  headers: { 'Authorization': 'Bearer al_live_YOUR_KEY' }
})
const { decisions, total } = await res.json()

{
  "decisions": [
    {
      "id": "dec_a1b2c3d4e5f6",
      "agent_id": "agent_xxx",
      "decision_type": "loan_approval",
      "input_summary": "Applicant: John Doe, credit score 720, income €45k",
      "output_summary": "Approved — €15,000 at 4.2% APR",
      "confidence": 0.91,
      "human_reviewed": true,
      "audit_hash": "sha256:4a7b3c9d2e1f8a...",
      "logged_at": "2025-01-01T12:00:00.000Z"
    }
  ],
  "total": 1,
  "limit": 20,
  "offset": 0
}

Log Decision

Record an agent decision in the immutable audit trail with a SHA-256 audit hash.

curl -X POST https://agentledgerhq.com/api/v1/decisions \
  -H "Authorization: Bearer al_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "agent_id": "agent_a1b2c3d4e5f6",
    "decision_type": "loan_approval",
    "input_summary": "Applicant credit profile hash: sha256:abc123...",
    "output_summary": "Decision: Approved. Loan amount: EUR 15000. Rate: 4.2%",
    "confidence": 0.87,
    "human_reviewed": true,
    "context": {
      "model_version": "2.1.0",
      "session_id": "ses_xyz"
    }
  }'

const res = await fetch('https://agentledgerhq.com/api/v1/decisions', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer al_live_YOUR_KEY',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    agent_id: 'agent_a1b2c3d4e5f6',
    decision_type: 'loan_approval',
    input_summary: 'Applicant credit profile hash: sha256:abc123...',
    output_summary: 'Decision: Approved. Loan amount: EUR 15000. Rate: 4.2%',
    confidence: 0.87,
    human_reviewed: true,
    context: { model_version: '2.1.0', session_id: 'ses_xyz' }
  })
})
const decision = await res.json() // 201 Created

{
  "id": "dec_a1b2c3d4e5f6",
  "agent_id": "agent_a1b2c3d4e5f6",
  "logged_at": "2025-01-01T12:00:00.000Z",
  "audit_hash": "sha256:4a7b3c9d2e1f8a6b5c4d3e2f1a0b9c8d7e6f5a4b3c2d1e0f..."
}

{
  "id": "dec_a1b2c3d4e5f6",
  "agent_id": "agent_a1b2c3d4e5f6",
  "logged_at": "2025-01-01T12:00:00.000Z",
  "audit_hash": "sha256:4a7b3c9d2e1f8a6b5c4d3e2f1a0b9c8d7e6f5a4b3c2d1e0f...",
  "warning": "High-risk agent decision with confidence > 0.9 and no human review. Consider mandatory human review before acting on this decision."
}

List API Keys

View all API keys associated with your account.

curl https://agentledgerhq.com/api/v1/keys \
  -H "Authorization: Bearer al_live_YOUR_KEY"

const res = await fetch('https://agentledgerhq.com/api/v1/keys', {
  headers: { 'Authorization': 'Bearer al_live_YOUR_KEY' }
})
const { keys } = await res.json()

{
  "keys": [
    {
      "id": "key_a1b2c3d4e5f6",
      "prefix": "al_live_xxxxxxxx...",
      "name": "Production",
      "created_at": "2025-01-01T12:00:00.000Z",
      "revoked_at": null
    }
  ]
}

Create API Key

Generate a new API key for your account.

curl -X POST https://agentledgerhq.com/api/v1/keys \
  -H "Authorization: Bearer al_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"name": "Staging Environment"}'

const res = await fetch('https://agentledgerhq.com/api/v1/keys', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer al_live_YOUR_KEY',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({ name: 'Staging Environment' })
})
const { key } = await res.json() // key.api_key shown only once

{
  "id": "key_a1b2c3d4e5f6",
  "api_key": "al_live_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx",
  "prefix": "al_live_xxxxxxxx...",
  "name": "Staging Environment",
  "created_at": "2025-01-01T12:00:00.000Z",
  "warning": "Store this key securely. It will not be shown again."
}

Revoke API Key

Permanently revoke an API key. This action cannot be undone.

curl -X DELETE https://agentledgerhq.com/api/v1/keys/key_a1b2c3d4e5f6 \
  -H "Authorization: Bearer al_live_YOUR_KEY"

const res = await fetch('https://agentledgerhq.com/api/v1/keys/key_a1b2c3d4e5f6', {
  method: 'DELETE',
  headers: { 'Authorization': 'Bearer al_live_YOUR_KEY' }
})
// 200 OK with {"message": "API key revoked successfully"}

{
  "message": "API key revoked successfully"
}

MCP Server

Agent Ledger exposes a native Model Context Protocol (MCP) server — the open standard for AI-to-tool communication. Any MCP-compatible AI agent can register itself, run EU AI Act scans, and log decisions autonomously — without writing a single REST call.

Setup & Authentication

The MCP server reuses your existing Agent Ledger API key. No separate credentials needed.

Authorization: Bearer al_live_your_api_key_here

curl -X POST https://agentledgerhq.com/api/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{}}'

{
  "jsonrpc": "2.0",
  "id": 1,
  "result": {
    "protocolVersion": "2024-11-05",
    "capabilities": { "tools": {} },
    "serverInfo": { "name": "Agent Ledger MCP", "version": "1.0.0" }
  }
}

Available Tools

8 tools available. Retrieve the full list with tools/list.

curl -X POST https://agentledgerhq.com/api/mcp \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer al_live_your_key" \
  -d '{
    "jsonrpc": "2.0",
    "id": 3,
    "method": "tools/call",
    "params": {
      "name": "run_risk_scan",
      "arguments": {
        "agent_id": "agent_abc123",
        "answers": {
          "q1_purpose": "credit_scoring",
          "q2_human_impact": "high",
          "q3_autonomy": "fully_autonomous",
          "q4_data_types": ["financial", "biometric"],
          "q5_human_oversight": false,
          "q6_affected_persons": ["loan_applicants"],
          "q7_sector": "finance"
        }
      }
    }
  }'

Client Configuration

Add Agent Ledger to your MCP client in seconds. Replace al_live_your_key with your actual API key.

{
  "mcpServers": {
    "agentledger": {
      "url": "https://agentledgerhq.com/api/mcp",
      "headers": {
        "Authorization": "Bearer al_live_your_key"
      }
    }
  }
}

{
  "plugins": {
    "mcp": {
      "servers": [
        {
          "name": "agentledger",
          "url": "https://agentledgerhq.com/api/mcp",
          "headers": {
            "Authorization": "Bearer al_live_your_key"
          }
        }
      ]
    }
  }
}

{
  "mcpServers": {
    "agentledger": {
      "url": "https://agentledgerhq.com/api/mcp",
      "headers": {
        "Authorization": "Bearer al_live_your_key"
      }
    }
  }
}

{
  "servers": {
    "agentledger": {
      "type": "http",
      "url": "https://agentledgerhq.com/api/mcp",
      "headers": {
        "Authorization": "Bearer al_live_your_key"
      }
    }
  }
}

Full Self-Compliance Example

An AI agent making itself EU AI Act compliant in 4 tool calls — no human required.

import json, httpx

MCP_URL = "https://agentledgerhq.com/api/mcp"
API_KEY = "al_live_your_key"

def mcp(method, params=None, req_id=1):
    r = httpx.post(MCP_URL, json={
        "jsonrpc": "2.0", "id": req_id,
        "method": method, "params": params or {}
    }, headers={"Authorization": f"Bearer {API_KEY}"})
    return r.json()["result"]

# Step 1: register this agent
reg = mcp("tools/call", {"name": "register_agent", "arguments": {
    "name": "CreditScore Bot v2",
    "description": "Automated credit scoring for loan applications",
    "sector": "finance"
}})
agent_id = json.loads(reg["content"][0]["text"])["agent_id"]
print(f"Registered: {agent_id}")

# Step 2: run EU AI Act risk scan
scan = mcp("tools/call", {"name": "run_risk_scan", "arguments": {
    "agent_id": agent_id,
    "answers": {
        "q1_purpose": "credit_scoring",
        "q2_human_impact": "high",
        "q3_autonomy": "semi_autonomous",
        "q4_data_types": ["financial"],
        "q5_human_oversight": True,
        "q7_sector": "credit"
    }
}})
result = json.loads(scan["content"][0]["text"])
print(f"Risk class: {result['risk_class']} (score: {result['risk_score']})")

# Step 3: log a decision
mcp("tools/call", {"name": "log_decision", "arguments": {
    "agent_id": agent_id,
    "decision_type": "credit_approval",
    "description": "Approved loan application #4821 based on credit score 712",
    "outcome": "approved",
    "confidence_score": 0.87
}})

# Step 4: get badge URL for README
badge = mcp("tools/call", {"name": "get_badge_url", "arguments": {"agent_id": agent_id}})
badge_data = json.loads(badge["content"][0]["text"])
print(f"Badge: {badge_data['markdown']}")