← All steps
Part B · Step 9

Build

Owner
Agent Builder.
Input
Identity + access in place.

Owner: Agent Builder. Input: Identity + access in place. Sub-steps:

  1. Work in dev environment only. Never start a new agent directly in prod.
  2. Use the approved stack (Step A4). If the build needs anything outside the stack, escalate to the CoE for an exception decision before adding it.
  3. Build the agent to the Agent Card spec. Every field in the card is a contract.
  4. Implement the three guardrail layers (framework.md §19):
    • Policy guardrails — already in the Agent Card.
    • Workflow guardrails — HITL gates, approval routing, RBAC at the orchestration layer.
    • Runtime guardrails — input validation, output validation, tool allowlists, schema checks, kill switch.
  5. Implement the five control mechanisms (framework.md §20):
    • Confidence thresholds per process.
    • HITL checkpoints at irreversible decision points.
    • Input validation + prompt-injection defense.
    • Deterministic workflow boundaries (orchestration owns the process, not the agent).
    • Least-privilege scoping at runtime.
  6. Wire in observability from day one (framework.md §24):
    • Every execution logs the full field list (timestamp, user, agent ID + version, prompt, output, tool calls, model + version, tokens, cost, policy checks, HITL events, latency, outcome, error).
    • Dashboards live before the agent ships, not after.
  7. Version everything — prompts, tool definitions, orchestration config — in source control. No production-only prompts.
  8. Write the runbook: what to do when this agent misbehaves, who to page, how to kill it.

Output / gate criteria: Agent runs in dev against test data. Logs flow. Guardrails fire in expected scenarios. Runbook exists.

Decision branches: none — this is the build.

Skip-this-step risk: "We'll add logging later" → never happens. Guardrails get bolted on as patches after the first incident. Production agent has no kill switch.