A2A & Agent Networks

GaaS as the governance control plane for multi-agent systems — covering the A2A Protocol v0.3, Agent Trust Registry, and AP2 Agentic Payment Governance.

When AI agents talk to each other, delegate tasks, or execute financial transactions autonomously, governance becomes non-negotiable. GaaS implements the Agent-to-Agent (A2A) Protocol v0.3 specification, acting as a governance proxy between agents: every inter-agent call passes through the same 5-stage pipeline as a human-facing action.

Agent Card

GaaS publishes a machine-readable agent card at a well-known URL so that other A2A-compatible systems can discover its capabilities, supported authentication methods, and governance endpoints.

GET /.well-known/agent.json

{
  "name": "GaaS Governance Agent",
  "version": "0.3.0",
  "description": "Governance as a Service — A2A-native control plane for autonomous agent networks",
  "url": "https://api.gaas.is",
  "capabilities": {
    "streaming": false,
    "pushNotifications": true,
    "stateTransitionHistory": true
  },
  "skills": [
    {
      "id": "governance.evaluate",
      "name": "Governance Evaluation",
      "description": "Evaluate an agent action intent against active governance policies"
    },
    {
      "id": "governance.proxy",
      "name": "A2A Governance Proxy",
      "description": "Route and govern inter-agent task delegation"
    }
  ],
  "authentication": {
    "schemes": ["Bearer", "OAuth2", "mTLS"]
  }
}

A2A Gateway

The A2A gateway exposes a JSON-RPC 2.0 endpoint that any A2A-compliant agent can call directly. Requests are automatically translated into GaaS intent declarations, run through the full governance pipeline, and translated back into A2A-compliant responses.

Supported JSON-RPC Methods

Example: A2A Task Submission

POST /a2a/jsonrpc
Content-Type: application/json
Authorization: Bearer <agent_token>

{
  "jsonrpc": "2.0",
  "id": "req_001",
  "method": "tasks/send",
  "params": {
    "id": "task_abc123",
    "message": {
      "role": "user",
      "parts": [
        {
          "type": "text",
          "text": "Transfer $2,500 to vendor account V-9912 for invoice #INV-2026-0044"
        }
      ]
    },
    "metadata": {
      "agent_id": "procurement_agent_v1",
      "delegated_by": "orchestrator_agent",
      "delegation_depth": 1
    }
  }
}

The gateway translates this into a TRANSACT intent, runs it through the governance pipeline (including AP2 mandate verification), and returns a JSON-RPC result containing the governance decision.

{
  "jsonrpc": "2.0",
  "id": "req_001",
  "result": {
    "id": "task_abc123",
    "status": {
      "state": "completed"
    },
    "artifacts": [
      {
        "type": "governance_decision",
        "decision": {
          "verdict": "approve",
          "risk_assessment": { "score": 0.18, "level": "LOW" },
          "ap2_context": {
            "mandate_id": "mnd_vendor_9912",
            "remaining_budget": 47500.00,
            "psd2_sca_required": false
          }
        }
      }
    ]
  }
}

Agent Trust Registry

The Agent Trust Registry is GaaS's decentralized reputation system for autonomous agents. Every agent that interacts with the platform accumulates a trust score based on governance outcomes, policy violations, and human review results.

Trust Score

Trust scores range from 0.0 (untrusted) to 1.0 (fully trusted). New agents start at 0.5 and build reputation through approved actions and positive human review outcomes. Violations, blocks, and escalations reduce the score.

A2A Authentication

The A2A gateway supports four authentication schemes, matching the A2A Protocol v0.3 specification:

• Bearer Token: Standard JWT token in the Authorization: Bearer header, validated against the agent registry
• OAuth 2.0: Client credentials flow for service-to-service authentication; access tokens are cached for their TTL
• mTLS: Mutual TLS for high-security environments; the client certificate's CN is mapped to an agent identity
• GaaS API Key: Standard X-API-Key header — compatible with existing integrations, automatically associated with an agent profile

Governance Proxy

For agents that delegate work to other agents, GaaS acts as a transparent governance proxy. The orchestrating agent registers the target agent endpoint; GaaS intercepts outbound calls, evaluates the intent, and forwards approved requests — all without modifying the agent's code.

# Register a downstream agent for governance proxying
POST /v1/agents
{
  "agent_id": "fulfillment_agent_v2",
  "endpoint": "https://fulfillment.internal/a2a",
  "capabilities": ["TRANSACT", "COMMUNICATE", "MODIFY"],
  "trust_level": "standard",
  "proxy_governance": true
}

Once registered, any A2A call your orchestrator makes to fulfillment_agent_v2 is automatically routed through the GaaS governance pipeline. If the action is approved, GaaS forwards the original request to the downstream agent and returns its response. If blocked or escalated, GaaS returns a governance decision and does not contact the downstream agent.

Cross-Org Policy Federation

Enterprise customers with multiple GaaS organizations (e.g., separate production and data-science orgs) can configure bidirectional policy federation. When an agent from org A calls into org B's governed scope, GaaS evaluates the intent against a merged policy set from both organizations.

Push Notifications

Agents can register webhook endpoints to receive real-time updates on task status changes, governance decisions, and escalation events. Push notifications are HMAC-SHA256 signed using the agent's registered secret.

POST /v1/a2a
{
  "jsonrpc": "2.0",
  "method": "tasks/pushNotification/set",
  "params": {
    "id": "task_abc123",
    "pushNotificationConfig": {
      "url": "https://your-agent.example.com/webhooks/gaas",
      "authentication": {
        "schemes": ["Bearer"],
        "credentials": "your_webhook_secret"
      }
    }
  }
}

AP2 Payment Governance

The Agentic Payment Protocol (AP2) extends GaaS with seven payment-specific policies that govern autonomous agent transactions. AP2 is designed around the emerging reality that AI agents will execute payments autonomously on behalf of users — and those payments require the same compliance safeguards as human-initiated transactions.

Tier 1 AP2 Policies (Fast-Fail)

These four policies run first and immediately block invalid or out-of-scope payment requests:

Tier 2 AP2 Policies (Regulatory)

These three policies enforce payment regulation compliance:

Submitting an AP2-governed Payment Intent

POST /v1/intents
X-API-Key: your_api_key

{
  "intent": {
    "agent": {
      "id": "procurement_agent_v1",
      "framework": "custom"
    },
    "action": {
      "type": "TRANSACT",
      "verb": "initiate_payment",
      "target": {
        "type": "ACCOUNT",
        "identifier": "vendor_account_V9912",
        "sensitivity": "REGULATED"
      }
    },
    "payload": {
      "summary": "Pay vendor invoice INV-2026-0044 for SaaS services",
      "content": {
        "amount_usd": 2500.00,
        "currency": "USD",
        "merchant_category": "7372",
        "mandate_id": "mnd_vendor_9912"
      }
    },
    "estimated_impact": {
      "reversible": false,
      "financial_exposure_usd": 2500.00,
      "regulatory_domains": ["PCI-DSS", "PSD2", "AML"]
    }
  }
}

AP2 Mandate Management

Mandates define the authorized scope for agent payments. A mandate specifies which agent may transact, on whose behalf, with which merchants, up to what limits, and for what period.

Mandate Schema

{
  "mandate_id": "mnd_vendor_9912",
  "agent_id": "procurement_agent_v1",
  "authorized_by": "user_cfo_01",
  "scope": {
    "merchant_categories": ["7372", "7371"],
    "max_single_transaction_usd": 5000.00,
    "daily_limit_usd": 10000.00,
    "monthly_limit_usd": 50000.00,
    "hnp_threshold_usd": 10000.00,
    "currency": "USD",
    "geographic_scope": ["US", "EU"]
  },
  "valid_from": "2026-01-01T00:00:00Z",
  "valid_until": "2026-12-31T23:59:59Z",
  "status": "active",
  "cumulative_spend_usd": 2500.00,
  "remaining_daily_budget_usd": 7500.00
}

Related Pages

• Connectors — A2A Registry and AP2 Mandate connectors
• Intent Declaration API — The core governance entry point
• Policy Library — Full policy catalog including AP2 and Tier 4 AI safety policies
• Webhooks — Event-driven notifications for governance decisions
• Authentication — API keys and A2A auth schemes