w-review — Claude Code skill: review changes and install
w-review — a skill for Claude Code checks changes for evidenced defects and risks. Run it on a diff, branch, PR, or working tree without asking it to edit code.
MIT 647 installs ★ 1 updated 2026-10-02
npx skills add blockmatic-icebox/basilic-skills --skill w-review -a claude-codeWhat w-review does
w-review assesses a change against its intended behavior and checks what the tests actually exercise. It examines affected code, its callers, and failure paths. Documentation is in scope when behavior changes.
The skill assigns review areas to 2–3 read-only explorer agents. Suspected defects need support from a concrete execution path, a failing case, or source evidence. Findings are ranked by impact and include a file and line, the trigger, the consequence, and a minimal remedy. Code inspection alone does not show that tests passed or approve a merge.
Automatic invocation is disabled, so the review must be requested explicitly.
What it helps with
- identify evidenced defects in changed code
- trace risks through failure paths
- check review claims against tests and intended behavior
Example prompts for the agent
- Use w-review to inspect the
feature/loginbranch and leave the code unchanged. - Review this PR for evidenced defects and risks without editing the working tree.
- Compare the diff with its intended behavior and report which tests support the findings.
How to install w-review
1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-review -a claude-code2Manually: put the files in place
~/.claude/skills/w-review/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/w-review/
1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-review -a codex2Manually: put the files in place
.agents/skills/w-review/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-review -a cursor2Manually: put the files in place
.cursor/skills/w-review/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install blockmatic-icebox/basilic-skills skills/workflow/w-review2Manually: put the files in place
.github/skills/w-review/SKILL.md1Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-review -a gemini-cli2Platform CLI
gemini skills install https://github.com/blockmatic-icebox/basilic-skills --path skills/workflow/w-reviewManually: put the files in place
.devin/skills/w-review/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add blockmatic-icebox/basilic-skills --skill w-review -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about w-review
How do I install w-review in Claude Code?
Run npx skills add blockmatic-icebox/basilic-skills --skill w-review -a claude-code. The skill file is placed at ~/.claude/skills/w-review/SKILL.md.
What does w-review do?
It reviews changes for defects and risks supported by code or execution evidence, taking tests into account. It leaves the change untouched unless fixes are requested too.
How does w-review validate a suspected defect?
It looks for a concrete execution path, a failing case, or evidence in the source. Findings are ordered by impact and include the affected location, trigger, consequence, and a minimal remedy.
Source text
SKILL.md · MIT
--- name: w-review description: Review a change for evidenced defects and risks without editing it. disable-model-invocation: true --- Use the specified diff, branch, PR, or working tree and its intended behavior. Remain read-only unless fixes are also requested. 1. Understand the trigger and expected result before judging the implementation. Read tests to learn which claims are actually exercised. 2. Check the in-scope [review dimensions](references/review-dimensions.md), including docs/README when behavior changed. Spawn 2–3 read-only explorers that split those dimensions. Concentrate on changed behavior, callers, and failure paths. Skip dimensions that are out of scope and say so. 3. Validate suspected defects with a concrete execution path, failing case, or source evidence. Do not invent timings or vulnerabilities from appearances. 4. Challenge verification claims. Order findings by impact with file/line, trigger, consequence, and smallest useful remedy. An inspection is not proof that tests passed and is not approval to merge.
© blockmatic-icebox · MIT · blockmatic-icebox/basilic-skills