What problem does it solve? Delivery and release decisions are often made on assumptions rather than evidence. This Skill replaces manual checklists and guesswork with grounded verdicts based on Dynatrace runtime telemetry, error budgets, SLOs, and deployment health data. ## Core Features & Use Cases - Three Gate Types: Design, Delivery, and Runtime gates, each with a defined verdict set (PASS, FAIL, ROLLBACK, LANE A/B/C) backed by DQL evidence queries. - Baseline Discovery: Ready-to-use DQL queries for service latency baselines, error budget status, deployment events, and host saturation. - Risk Tiering: A Tier 0/1/2 assessment framework that matches lifecycle depth to change risk, so low-risk changes skip unnecessary gates. - Post-Deploy Verification: Pre/post latency comparison, new-problem detection, feature flag checks, and dtctl wait query rollout assertions. - Use Case: Before shipping a hot-path change, run the design gate to collect baselines and map dependencies, then after deployment compare p90 latency against the pre-deploy baseline and classify any regression into infra, code, or design lanes. ## Quick Start Ask the assistant to run an evidence-based delivery gate for a service by collecting its Dynatrace baseline and evaluating whether the change is safe to ship.
