ada-code-quality-report-verification

Verifies code quality report claims against actual source code and reproducible tool output.

Updated Jul 23, 2026
One-click install
npx skills add https://github.com/wubing7755/Ada --skill ada-code-quality-report-verification-wubing7755
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ada-code-quality-report-verification
Source: https://github.com/wubing7755/Ada/tree/main/skills/software-development/ada-code-quality-report-verification
Command: npx skills add https://github.com/wubing7755/Ada --skill ada-code-quality-report-verification-wubing7755

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Code quality reports can contain fabricated findings, wrong counts, stale statistics, or unsupported pass/fail verdicts. This Skill independently re-checks every claim against the actual source code before you act on a report's recommendations. ## Core Features & Use Cases - Issue Sampling (Mode A): Spot-check 5 issues from a report's P0/P1/P2 lists and classify each as confirmed, partially correct, or false positive. - Full Statistical Audit (Mode B): Exhaustively verify every quantitative claim — file counts, line counts, coverage percentages, test method counts — with a 7-point self-consistency protocol. - Phase Pass/Fail Review (Mode C): Fail-closed review of an implementation phase against SRS/design acceptance criteria, returning a compact JSON-like verdict. - Use Case: Before sending a quality report to your team, ask the agent to independently verify its statistics and issue claims; it re-runs tools like dotnet format and grep to confirm or refute each finding. ## Quick Start Ask the agent to independently verify the issues and statistics in your existing code quality report against the actual source code.

Frequently Asked Questions about ada-code-quality-report-verification

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

FAQPage Schema
How do I verify a code quality report against source code?

Sample issues from the report's priority lists, navigate to each claimed file:line, read the actual code, and re-run cited tools like dotnet format or grep counts. Classify each issue as confirmed, partially correct, or false positive with concrete evidence.

How to audit statistics in a code quality report for consistency?

Independently count files and lines with find and wc -l excluding obj/ directories, sum the report's own subdirectory breakdowns to check internal consistency, and grep test files to verify coverage claims. Compare each actual value against the reported value.

When should I use report verification instead of code quality analysis?

Use report verification only when a quality report already exists and you need to validate its claims. To generate a new report from scratch, use a code quality analysis skill instead; verification never creates reports.

Can it review whether an implementation phase passed acceptance criteria?

Yes, Mode C performs fail-closed reviews: it reads SRS/design acceptance criteria first, inspects changed source and focused tests, runs test slices and build gates, and returns a compact passed:true/false verdict with evidence.

Why do coverage counts in reports often turn out wrong?

Reports frequently miss aggregated test files that contain multiple test classes, or exclude interfaces and DTOs from class counts. Always grep test files for class references rather than trusting the report's coverage table.