What problem does it solve? Engineering teams track token costs but lack visibility into actual delivery performance — how fast and how reliably they ship. This Skill computes the DORA four keys plus reliability and Flow Framework metrics from data the harness already records, turning raw timestamps and incident events into banded performance reports without manual data collection. ## Core Features & Use Cases - DORA metrics report: Computes deployment frequency, lead time for change, change failure rate, MTTR, and reliability, banded against the 2024 DORA State of DevOps clusters (elite/high/medium/low). - Flow Framework metrics: Calculates flow efficiency (active time ÷ total flow time), flow time, and flow load from per-phase cost-ledger checkpoints. - Event-driven delivery ledger: The delivery-record script upserts deploy, incident-start, incident-resolve, and SLO events into a JSONL ledger so failure-dependent keys reflect real events rather than hand-edited data. - Anti-fabrication guardrails: Withholds any metric whose input signal is missing (e.g., prints "needs incident linkage") instead of reporting a fabricated zero, and a reviewer gate validates the report before stakeholder use. - Use Case: After merging a feature branch, run the report to see your lead time band and change failure rate for the last 30 days, then feed MTTR trends back into incident-response and error-budget policy. ## Quick Start Run the delivery metrics report for the last 30 days and show my DORA bands and flow efficiency.
