Process & tooling / L2 detail
Lifecycle management
Linked Level 3 activities
Level 3
Pre-deployment evaluation (capability and safety)
Treat evaluation as a product and tooling capability - datasets, harnesses, trajectory evals, red teaming, acceptance thresholds, and regression gates, not a light-touch checklist
Open Level 3 detail
Level 3
Change and release management
Manage releases as bundles (agents + model + tools/connectors + policies + prompts + memory/config), with controlled rollout and rollback
Open Level 3 detail
Level 3
Live tuning, configuration, and update management
Define how live agents are tuned and updated - configuration management, policy updates, safe rollout, and rollback - rather than informal prompt tweaking
Open Level 3 detail
Level 3
Retirement and decommissioning
Retire agents safely by removing tool access, archiving evidence, and migrating workflows
Open Level 3 detail
Level 3
Post-incident review and remediation workflow
Run PIRs focused on autonomy breakdowns, control failures, and tool-chain issues, with tracked remediation
Open Level 3 detail