harness-quality-snapshot

Create milestone quality scorecards citing gate report evidence and trends.

Updated Jul 29, 2026
One-click install
npx skills add https://github.com/MaiconGambini/opencode-harness-guide --skill harness-quality-snapshot-maicongambini
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-quality-snapshot
Source: https://github.com/MaiconGambini/opencode-harness-guide/tree/main/skills/harness-quality-snapshot
Command: npx skills add https://github.com/MaiconGambini/opencode-harness-guide --skill harness-quality-snapshot-maicongambini

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams running continuous quality gates lack a per-milestone view of project health: gate reports are per-change and binary, so no single run shows direction over time or which risk to act on next. ## Core Features & Use Cases - Milestone scorecard: Produces a quality snapshot with an overall score out of 10, an evidence table, top risks, and the next improvement, stored in docs/harness/quality-document.md. - Gate-cited dimensions: Reads docs/harness/quality/*.json gate reports for mutation kill ratio, line coverage, regression suite, boundary violations, and measurement coverage instead of re-deriving metrics. - Trend interpretation: Highlights direction over time (e.g., mutation 62% to 69% across milestones) and flags unavailable metrics as confidence risks. - Use Case: Before a release or capstone review, generate the snapshot to summarize build health, test strength, architecture boundaries, and cleanup priorities with cited evidence. ## Quick Start Create a quality snapshot for this release milestone using the latest gate reports in docs/harness/quality and write it to docs/harness/quality-document.md.

Frequently Asked Questions about harness-quality-snapshot

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a milestone quality scorecard for a project?

Run the snapshot at a release, capstone, or major sprint review. It writes a scorecard to docs/harness/quality-document.md with an overall score out of 10, an evidence table, top risks, and the next improvement, citing the latest gate report for each measured dimension.

What is the difference between a quality gate and a quality snapshot?

The gate is per-change and binary, producing pass/fail numbers. The snapshot is per-milestone and qualitative, interpreting the gate's numbers as trends over time and identifying which risk to act on next.

Should I re-run coverage or mutation analysis for the snapshot?

No. The snapshot cites the latest gate report in docs/harness/quality/*.json rather than re-deriving coverage, mutation, complexity, or boundary findings by inspection. It quotes the gate's numbers directly.

When should I not use a milestone quality snapshot?

Do not use it for every small task; it is intended for milestones only. It also requires existing gate reports, so without per-change gate data there is no evidence base to cite.

How are unavailable metrics handled in a quality scorecard?

An unavailable metric lowers confidence in the whole snapshot and must be stated explicitly rather than scored around. A gate measuring only a few of the expected rows is itself flagged as a risk.