#601

Globales Ranking · von 601 Skills

critique AI Agent Skill

Quellcode ansehen: b-open-io/prompts

Critical

Installation

npx skills add b-open-io/prompts --skill critique

54

Installationen

Review git diffs with syntax highlighting, split view, and word-level diff. Show users what code was changed during a session using terminal UI or web preview.

Prerequisites

Critique requires Bun (does not work with Node.js):

# Run directly (no install needed)
bunx critique

# Or install globally
bun install -g critique

Core Commands

View Current Changes

# Unstaged changes (includes untracked files)
critique

# Staged changes only
critique --staged

# Watch mode - auto-refresh on file changes
critique --watch

View Commits

# Last commit
critique HEAD

# Specific commit
critique --commit HEAD~1
critique --commit abc1234

# Combined changes from last N commits
critique HEAD~3 HEAD

Compare Branches (PR-style)

# What feature-branch added vs main
critique main feature-branch

# What current branch added vs main
critique main HEAD

Filter Files

# Filter by glob pattern
critique --filter "src/**/*.ts"
critique --filter "src/**/*.ts" --filter "lib/**/*.js"

Web Preview

Generate an HTML preview viewable in a browser - ideal for sharing or detailed review:

# Generate web preview
critique --web

# Generate and open in browser
critique --web --open

# Web preview for commits
critique HEAD --web --open
critique main HEAD --web --open

The web preview provides:

  • Full syntax highlighting
  • Split view with before/after
  • Shareable HTML output
  • Better for large diffs

AI-Powered Review

Generate AI explanations of code changes. Best for reviewing AI-generated changes:

# Review unstaged changes (uses OpenCode by default)
critique review

# Use Claude Code as the AI backend
critique review --agent claude

# Review specific commits
critique review HEAD
critique review --commit HEAD~1

# Review branch diff (like a PR)
critique review main HEAD

# Include session context for better explanations
critique review --agent claude --session <session-id>

# Generate web preview of AI review
critique review --web
critique review --web --open

AI review features:

  • Orders hunks in logical reading order
  • Splits large hunks into digestible chunks
  • Explains changes with diagrams and tables
  • Groups related changes across files

Integrations

Git Difftool

Configure as default git difftool:

git config --global diff.tool critique
git config --global difftool.critique.cmd 'critique difftool "$LOCAL" "$REMOTE"'

Then use: git difftool HEAD~1

Lazygit

Add to ~/.config/lazygit/config.yml:

git:
  paging:
    pager: critique --stdin

Running from Claude Code

Since the TUI runs inside the Bash tool (hidden/folded), use one of these approaches:

Option 1: Split Pane (macOS with iTerm2, Recommended)

Open critique in a split pane that auto-closes when done:

# Horizontal split (top/bottom)
/path/to/skills/critique/scripts/open-critique-pane.sh /path/to/repo -h

# Vertical split (side by side)
/path/to/skills/critique/scripts/open-critique-pane.sh /path/to/repo -v

Option 2: New Terminal Tab (macOS with iTerm2)

Open critique in a new tab:

/path/to/skills/critique/scripts/open-critique.sh /path/to/repo

Option 3: Web Preview (Cross-platform)

Opens in browser - works everywhere:

bunx critique --web --open

Recommended Usage After Claude Work

When a user asks to see what changed after Claude has made modifications:

  1. New terminal tab (macOS): Use osascript to open critique in iTerm
  2. Web preview: Run critique --web --open for browser view
  3. AI-explained review: Run critique review --agent claude --web --open
  4. Compare to starting point: Run critique HEAD~N HEAD --web --open

Navigation Keys (Terminal UI)

Key Action
j/k or ↓/↑ Navigate lines
h/l or ←/→ Switch files
Tab Toggle split/unified view
q Quit
? Help

Installationen

Installationen 54
Globales Ranking #601 von 601

Sicherheitsprüfung

ath High
socket Safe
Warnungen: 0 Bewertung: 90
snyk Low
EU EU-Hosted Inference API

Power your AI Agents with the best open-source models.

Drop-in OpenAI-compatible API. No data leaves Europe.

Explore Inference API

GLM

GLM 5

$1.00 / $3.20

per M tokens

Kimi

Kimi K2.5

$0.60 / $2.80

per M tokens

MiniMax

MiniMax M2.5

$0.30 / $1.20

per M tokens

Qwen

Qwen3.5 122B

$0.40 / $3.00

per M tokens

So verwenden Sie diesen Skill

1

Install critique by running npx skills add b-open-io/prompts --skill critique in your project directory. Führen Sie den obigen Installationsbefehl in Ihrem Projektverzeichnis aus. Die Skill-Datei wird von GitHub heruntergeladen und in Ihrem Projekt platziert.

2

Keine Konfiguration erforderlich. Ihr KI-Agent (Claude Code, Cursor, Windsurf usw.) erkennt installierte Skills automatisch und nutzt sie als Kontext bei der Code-Generierung.

3

Der Skill verbessert das Verständnis Ihres Agenten für critique, und hilft ihm, etablierte Muster zu befolgen, häufige Fehler zu vermeiden und produktionsreifen Code zu erzeugen.

Was Sie erhalten

Skills sind Klartext-Anweisungsdateien — kein ausführbarer Code. Sie kodieren Expertenwissen über Frameworks, Sprachen oder Tools, das Ihr KI-Agent liest, um seine Ausgabe zu verbessern. Das bedeutet null Laufzeit-Overhead, keine Abhängigkeitskonflikte und volle Transparenz: Sie können jede Anweisung vor der Installation lesen und prüfen.

Kompatibilität

Dieser Skill funktioniert mit jedem KI-Coding-Agenten, der das skills.sh-Format unterstützt, einschließlich Claude Code (Anthropic), Cursor, Windsurf, Cline, Aider und anderen Tools, die projektbezogene Kontextdateien lesen. Skills sind auf Transportebene framework-agnostisch — der Inhalt bestimmt, für welche Sprache oder welches Framework er gilt.

Data sourced from the skills.sh registry and GitHub. Install counts and security audits are updated regularly.

EU Made in Europe

Chat with 100+ AI Models in one App.

Use Claude, ChatGPT, Gemini alongside with EU-Hosted Models like Deepseek, GLM-5, Kimi K2.5 and many more.

Start for free View pricing

App herunterladen:

Laden im App Store Jetzt bei Google Play
Kundensupport