ICMD — Agent Readiness & Rollout Checklist (30 Days) Use this to ship one agent workflow safely and measurably. 1) Choose the Workflow (Day 1–2) - Pick ONE workflow with high repetition and a clear owner. - Define the System of Record (SoR): where truth lives (e.g., Zendesk, Salesforce, Jira). - Write a one-sentence outcome: “Agent resolves billing tickets that qualify for a templated credit under a defined threshold.” - Establish baseline metrics: cycle time, error rate, and weekly volume. 2) Define “Done” + Verification (Day 3–5) - Define objective success checks (system state change, test pass, or explicit approval). - Define non-goals (what the agent must not do). - Create a failure taxonomy: retrieval miss, policy violation, tool error, ambiguous request. - Decide escalation rules: when to ask a human and who owns the queue. 3) Instrumentation (Week 1) - Log at 3 levels: session (intent), plan (steps), execution (tool calls + diffs). - Track: verified outcome rate, cost per verified outcome, rollback rate, and completion time by workflow. - Add spend controls immediately: per-task cap and daily workspace budget. 4) Ship Autonomy Levels (Weeks 2–4) - Week 2: Suggest mode only (drafts). Measure acceptance and how much humans edit. - Week 3: Queue mode (preview + approve). Add citations and diff views. - Week 4: Constrained execute for low-risk actions (threshold-based, with approvals above the threshold). 5) Governance & Admin (Parallel) - Permissions: least privilege; read-only by default; time-bound tokens. - Audit logs: who triggered the run, what tools were called, what changed, timestamps. - Policy controls: thresholds, approvals, and environment separation (prod vs staging). - Rollback/quarantine: ability to revert changes and pause the agent. 6) Promotion Criteria (End of Month) - Promote autonomy only after sustained performance across: - Verified outcomes that can be audited - Rollback coverage for write actions - Task cost that fits your margin model - No severe incident without a documented fix plus a regression test 7) Enterprise Expansion Readiness - SSO (Okta/Entra/Google), role management, retention controls, and admin dashboards. - Document security posture (SOC 2 plan, encryption, incident response). - Create an internal “agent runbook”: kill switch, on-call, escalation, and postmortem template. If you can’t verify outcomes, you can’t automate responsibly. Start narrow, measure relentlessly, and expand autonomy only after reliability and governance are real—not implied.