semi-formal-code-review

Analyze code changes to produce evidence-backed reviews with file:line citations.

Updated Feb 25, 2026
One-click install
npx skills add https://github.com/mikewolfd/formspec --skill semi-formal-code-review-mikewolfd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: semi-formal-code-review
Source: https://github.com/mikewolfd/formspec/tree/main/.claude-plugin/skills/semi-formal-code-review
Command: npx skills add https://github.com/mikewolfd/formspec --skill semi-formal-code-review-mikewolfd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Helps reviewers produce rigorous, evidence-backed assessments of code changes by requiring concrete file:line citations and traceable code paths.

Core Features & Use Cases

  • Evidence-first review comments with concrete file:line support.
  • Reading changed implementations, not just diff summaries.
  • Tracing important callers and callees for changed code paths.
  • Inspecting relevant tests to verify coverage and identify gaps.

Quick Start

Provide a PR diff and ask the agent to generate a findings-first, evidence-backed review.

Frequently Asked Questions about semi-formal-code-review

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

FAQPage Schema
How do I generate an evidence-backed code review for a PR diff?

To generate an evidence-backed code review, provide a PR diff and the agent analyzes changed implementations to produce structured findings with concrete file:line citations. It traces callers, data flows, and tests to flag un-tested paths.

What is evidence-first code review and how does it trace changes?

Evidence-first code review is a structured assessment requiring concrete file:line citations for all findings. It traces important callers and callees for changed code paths and inspects relevant tests to verify coverage and identify gaps.

Can I use static analysis to find un-tested paths in a codebase patch?

Yes, you can use static analysis on a codebase patch to find un-tested paths. The agent inspects relevant tests to verify coverage, reading the changed implementations rather than just diff summaries to flag testing gaps.

Does traceability in code review help with diff analysis across codebases?

Traceability in code review helps with diff analysis across codebases by ensuring all findings cite concrete evidence. It applies structured reviews to PRs, diffs, and patches, tracing data flows and callers to validate changes.

What's the best way to verify test coverage for changed code paths?

The best way to verify test coverage for changed code paths is inspecting relevant tests alongside tracing callers and callees. This evidence-first approach flags un-tested paths and gaps by reading the actual changed implementations.

Why does my code review lack traceable findings when analyzing patches?

Your code review lacks traceable findings when it relies on diff summaries instead of reading changed implementations. Applying evidence-first reviews with concrete file:line citations ensures all findings trace callers, data flows, and tests.