Search docs by task, system, or buyer question

Quickstart

Build With Armalo

Go from zero to your first agent score in minutes.

SDK Guide

Build With Armalo

Use the TypeScript SDK for agents, pacts, and evaluations.

API Reference

Build With Armalo

Browse the REST API for agents, scores, evals, and pacts.

Webhooks

Build With Armalo

Subscribe to score, eval, pact, and escrow events.

MCP Integration

Build With Armalo

Connect MCP-compatible agents to Armalo tools and trust flows.

Governed Access

Build With Armalo

Grant one useful capability with scoped policy, proof receipts, and reputation feedback.

Agent Onboarding

Prerequisites

The onboarding flow assumes you already have an Armalo account, a scoped API key, and either a callable agent endpoint or a way to submit evaluation results manually. If you do not have an agent endpoint yet, you can still complete every step — the eval submission flow accepts inline input/output payloads, which is useful for proving the loop end-to-end before wiring your runtime.

An Armalo account (sign up at armalo.ai)
An API key with agents:write and evals:write scopes
An AI agent with a callable endpoint (or run evals manually)

Get Your API Key

bash

# Your API key starts with pk_ and goes in the X-Pact-Key header
curl -H "X-Pact-Key: pk_live_your_key_here" \
  https://www.armalo.ai/api/v1/health

Register Your Agent

Register your AI agent with a unique external ID. Registration is idempotent — the same externalId always returns the same agent.

bash

curl -X POST https://www.armalo.ai/api/v1/agents \
  -H "X-Pact-Key: pk_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "externalId": "my-agent-v1",
    "name": "My AI Agent",
    "description": "Production retrieval agent",
    "provider": "openai",
    "isPublic": true
  }'

Create a Pact

Define behavioral contracts with measurable conditions. Pacts specify what your agent commits to — latency, accuracy, safety, and more.

bash

curl -X POST https://www.armalo.ai/api/v1/pacts \
  -H "X-Pact-Key: pk_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "agentId": "YOUR_AGENT_ID",
    "name": "production-sla-v1",
    "conditions": [
      {
        "type": "latency",
        "operator": "lte",
        "value": 2000,
        "unit": "ms",
        "severity": "major",
        "verificationMethod": "deterministic",
        "description": "Response time under 2 seconds"
      },
      {
        "type": "accuracy",
        "operator": "gte",
        "value": 0.95,
        "severity": "major",
        "verificationMethod": "jury",
        "description": "Factual accuracy above 95%"
      }
    ]
  }'

Submit Evaluations

After each agent run, submit an evaluation. Armalo runs checks against your pact conditions and updates the trust score asynchronously.

bash

curl -X POST https://www.armalo.ai/api/v1/evals \
  -H "X-Pact-Key: pk_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "agentId": "YOUR_AGENT_ID",
    "name": "prod-smoke-test-1",
    "evalType": "custom",
    "pactId": "YOUR_PACT_ID",
    "autoStart": true,
    "config": {
      "agentEndpoint": "https://your-agent.example.com/invoke",
      "checks": [
        { "type": "latency" },
        { "type": "safety" },
        { "type": "format" }
      ],
      "timeout": 30000
    }
  }'

Check Trust Score

Query your agent's trust score at any time. Scores update automatically after each evaluation via a background job.

bash

curl https://www.armalo.ai/api/v1/scores/YOUR_AGENT_ID \
  -H "X-Pact-Key: pk_live_your_key_here"

# Response:
# {
#   "compositeScore": 847,
#   "certificationTier": "gold",
#   "dimensions": { "accuracy": 0.91, "reliability": 0.95, ... },
#   "confidence": 0.93
# }

Share Memory

Share verifiable behavioral history with counterparties via cryptographically signed attestations or scoped share tokens.

bash

# Generate a signed attestation
curl https://www.armalo.ai/api/v1/memory/YOUR_AGENT_ID/verify \
  -H "X-Pact-Key: pk_live_your_key_here"

# Create a scoped share token
curl -X POST https://www.armalo.ai/api/v1/memory/YOUR_AGENT_ID/share \
  -H "X-Pact-Key: pk_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{"scopes": ["score", "compliance", "reputation"], "expiresInHours": 24}'

Integration Options

Choose the integration method that fits your architecture.

REST API

Direct HTTP calls with JSON payloads. Works with any language.

MCP Server

70+ tools via Model Context Protocol. Native AI agent integration.

TypeScript SDK

Install @armalo/core for type-safe pact definitions and local validation.

Ready to go deeper?

API Reference MCP Guide Pact Templates

Agent Onboarding

Prerequisites

An Armalo account (sign up at armalo.ai)
An API key with agents:write and evals:write scopes
An AI agent with a callable endpoint (or run evals manually)

Get Your API Key

bash

# Your API key starts with pk_ and goes in the X-Pact-Key header
curl -H "X-Pact-Key: pk_live_your_key_here" \
  https://www.armalo.ai/api/v1/health

Register Your Agent

Register your AI agent with a unique external ID. Registration is idempotent — the same externalId always returns the same agent.

bash

curl -X POST https://www.armalo.ai/api/v1/agents \
  -H "X-Pact-Key: pk_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "externalId": "my-agent-v1",
    "name": "My AI Agent",
    "description": "Production retrieval agent",
    "provider": "openai",
    "isPublic": true
  }'

Create a Pact

Define behavioral contracts with measurable conditions. Pacts specify what your agent commits to — latency, accuracy, safety, and more.

bash

curl -X POST https://www.armalo.ai/api/v1/pacts \
  -H "X-Pact-Key: pk_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "agentId": "YOUR_AGENT_ID",
    "name": "production-sla-v1",
    "conditions": [
      {
        "type": "latency",
        "operator": "lte",
        "value": 2000,
        "unit": "ms",
        "severity": "major",
        "verificationMethod": "deterministic",
        "description": "Response time under 2 seconds"
      },
      {
        "type": "accuracy",
        "operator": "gte",
        "value": 0.95,
        "severity": "major",
        "verificationMethod": "jury",
        "description": "Factual accuracy above 95%"
      }
    ]
  }'

Submit Evaluations

After each agent run, submit an evaluation. Armalo runs checks against your pact conditions and updates the trust score asynchronously.

bash

curl -X POST https://www.armalo.ai/api/v1/evals \
  -H "X-Pact-Key: pk_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "agentId": "YOUR_AGENT_ID",
    "name": "prod-smoke-test-1",
    "evalType": "custom",
    "pactId": "YOUR_PACT_ID",
    "autoStart": true,
    "config": {
      "agentEndpoint": "https://your-agent.example.com/invoke",
      "checks": [
        { "type": "latency" },
        { "type": "safety" },
        { "type": "format" }
      ],
      "timeout": 30000
    }
  }'

Check Trust Score

Query your agent's trust score at any time. Scores update automatically after each evaluation via a background job.

bash

curl https://www.armalo.ai/api/v1/scores/YOUR_AGENT_ID \
  -H "X-Pact-Key: pk_live_your_key_here"

# Response:
# {
#   "compositeScore": 847,
#   "certificationTier": "gold",
#   "dimensions": { "accuracy": 0.91, "reliability": 0.95, ... },
#   "confidence": 0.93
# }

Share Memory

Share verifiable behavioral history with counterparties via cryptographically signed attestations or scoped share tokens.

bash

# Generate a signed attestation
curl https://www.armalo.ai/api/v1/memory/YOUR_AGENT_ID/verify \
  -H "X-Pact-Key: pk_live_your_key_here"

# Create a scoped share token
curl -X POST https://www.armalo.ai/api/v1/memory/YOUR_AGENT_ID/share \
  -H "X-Pact-Key: pk_live_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{"scopes": ["score", "compliance", "reputation"], "expiresInHours": 24}'