skills/critique/SKILL.md
Git diff viewer. Renders diffs as web pages, images, and PDFs with syntax highlighting. Use this skill when working with critique for showing diffs, generating diff URLs, or selective hunk staging.
npx skillsauth add remorses/critique critiqueInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Git diff viewer that renders diffs as web pages, images, and PDFs with syntax highlighting.
Agents running in headless environments (kimaki on Discord, openclaw on Slack/Telegram) have no terminal to show diffs. critique uploads diffs to critique.work and returns a shareable URL you can paste into chat. Users click the link and see a syntax-highlighted split-view diff with mobile support and dark/light mode — no install needed.
Always run critique --help first to see the latest flags and commands. The help output is the source of truth.
Always pass a title to describe what the diff contains.
# Working tree changes
critique --web "Add retry logic to database connections"
# Staged changes
critique --staged --web "Refactor auth middleware"
# Branch diff (three-dot: changes since diverging from base)
critique main...HEAD --web "Feature branch changes"
critique main...feature-branch --web "Compare branches"
# Last N commits
critique HEAD~3 --web "Recent changes"
# Specific commit
critique --commit HEAD --web "Latest commit"
critique --commit abc1234 --web "Fix race condition"
# Filter to specific files
critique --web "API changes" --filter "src/api.ts" --filter "src/utils.ts"
# JSON output for programmatic use (returns {url, id, files})
critique --web "Deploy changes" --json
Share the returned URL with the user so they can see the diff.
critique --pdf # working tree to PDF
critique --staged --pdf # staged changes
critique main...HEAD --pdf # branch diff
critique --commit HEAD --pdf # single commit
critique --pdf output.pdf # custom filename
critique --pdf --pdf-page-size a4-portrait # page size options
critique main...HEAD --pdf --open # open in viewer
critique --image # renders to /tmp as WebP
critique main...HEAD --image # branch diff as images
When multiple agents work on the same repo, each agent should only commit its own changes. critique hunks lets you stage individual hunks instead of whole files — like a scriptable git add -p.
# List hunks with stable IDs
critique hunks list
critique hunks list --filter "src/**/*.ts"
# Stage specific hunks by ID
critique hunks add 'src/main.ts:@-10,6+10,7'
critique hunks add 'src/main.ts:@-10,6+10,7' 'src/utils.ts:@-5,3+5,4'
Hunk ID format: file:@-oldStart,oldLines+newStart,newLines — derived from the @@ diff header, stable across runs.
Typical workflow:
critique hunks list # see all unstaged hunks
critique hunks add 'file:@-10,6+10,7' # stage only your hunks
git commit -m "your changes" # commit separately
Every --web upload also stores the raw unified diff. Append .patch to any critique URL to get it:
# View the raw patch
curl https://critique.work/v/<id>.patch
# Apply the patch to current repo
curl -s https://critique.work/v/<id>.patch | git apply
# Reverse the patch (undo the changes)
curl -s https://critique.work/v/<id>.patch | git apply --reverse
Useful when an agent shares a critique URL and you want to programmatically apply or revert those changes.
bunx critique or global critique--web URLs expire after 7 days (content-hashed, same diff = same URL)development
Maintainer-only workflow for handling GitHub Secret Scanning alerts on OpenClaw. Use when Codex needs to triage, redact, clean up, and resolve secret leakage found in issue comments, issue bodies, PR comments, or other GitHub content.
development
Maintainer workflow for OpenClaw releases, prereleases, changelog release notes, and publish validation. Use when Codex needs to prepare or verify stable or beta release steps, align version naming, assemble release notes, check release auth requirements, or validate publish-time commands and artifacts.
development
Run, watch, debug, and extend OpenClaw QA testing with qa-lab and qa-channel. Use when Codex needs to execute the repo-backed QA suite, inspect live QA artifacts, debug failing scenarios, add new QA scenarios, or explain the OpenClaw QA workflow. Prefer the live OpenAI lane with regular openai/gpt-5.4 in fast mode; do not use gpt-5.4-pro or gpt-5.4-mini unless the user explicitly overrides that policy.
development
End-to-end Parallels smoke, upgrade, and rerun workflow for OpenClaw across macOS, Windows, and Linux guests. Use when Codex needs to run, rerun, debug, or interpret VM-based install, onboarding, gateway smoke tests, latest-release-to-main upgrade checks, fresh snapshot retests, or optional Discord roundtrip verification under Parallels.