Anthropic's official code-review plugin for Claude Code: runs a structured review of a diff or pull request for correctness, standards and spec adherence and reports findings with locations and severity.
skillgild install code-review
License: Apache-2.0 · Created by Anthropic · The project lists support for Claude Code
A community skill: it installs from its own repository and your agent runs the project's instructions locally with your own model. SkillGild lists it and credits the author; nothing runs on SkillGild.
With the SkillGild CLI (needs git). It clones anthropics/claude-plugins-official and copies the skill folder into the right place for your agent; use --agent codex, cursor or gemini-cli for other clients.
skillgild install code-review --agent claude-codeClaude Code plugin marketplace.
/plugin install code-review@claude-plugins-officialManual copy.
Install from the official marketplace; Claude Code adds it automaticallyOr copy the folder yourself from anthropics/claude-plugins-official (folder plugins/code-review) into ~/.claude/skills/code-review/ for Claude Code, ~/.agents/skills/ for Codex, ~/.cursor/skills/ for Cursor or ~/.gemini/skills/ for Gemini CLI, then start a new session.
Requests that trigger this skill once it is installed in your agent.
Review the current diff
Code-review PR #142 against the spec
Find correctness bugs in the payments branch
Run skillgild install code-review --agent claude-code: the SkillGild CLI clones anthropics/claude-plugins-official and copies the skill folder into ~/.claude/skills/code-review. You can also use the project's own commands shown above, or copy the folder yourself. Use --agent codex, cursor or gemini-cli for other clients.
Yes. The upstream repository is published under the Apache-2.0 license by Anthropic. SkillGild lists and credits it and installs it from source; it does not host or run it.
The project lists Claude Code. It is a plain Agent Skills folder (SKILL.md plus supporting files), so any client that reads that format can load it.
No. Code Review (Anthropic) is a community skill: your agent runs its instructions locally with your own model, and SkillGild lists, credits and installs it. Hosted SkillGild skills are marked differently and run through the SkillGild MCP server.
Setup steps and worked examples for this kind of skill.
Karpathy's observations on how LLMs fail at coding, codified into four rules the agent follows: think before coding (state assumptions, ask instead of guessing), simplicity first, surgical changes, and goal-driven execution.
A large community collection: 380+ skills, 30+ agents and 70+ commands for Claude Code, Codex, Gemini CLI, Cursor and other agents, organised by job (engineering, marketing, product, compliance, C-level advisory, research, finance, operations, daily productivity).
Anthropic's official plugin that adds commit, push and pull-request commands to Claude Code: /commit-commands:commit writes a conventional message from the diff, pushes, and can open the PR.
Makes the agent interview you about a plan or design, one question at a time with a recommended answer attached, until every branch of the decision tree is resolved and you both share the same understanding.
A context-engineering framework that runs Claude Code through a disciplined phase loop (new-project or onboard, plan-phase, execute-phase in parallel waves, verify-work, ship) with research and execution in fresh-context subagents so the main session never rots.
Compacts a long conversation into a handoff document another agent (or a fresh session) can continue from: the goal, decisions made, current state, open questions and the exact next step.
Anthropic's skill for building MCP servers: it walks the agent through tool design, schemas, transports and testing so the server it generates follows the Model Context Protocol correctly and works in Claude Code, Cursor and Codex.
Turn a feature request into an OpenSpec change before any code is written: proposal, delta specs, design and implementation tasks, grounded in the project's existing specs.
The lazy senior developer ruleset: before writing code the agent climbs a decision ladder (is it needed, does something reusable exist, is there a built-in) and only then writes the minimum.
Vercel Engineering's React and Next.js performance rules as a skill: 40+ rules in eight categories, prioritised by impact, covering waterfalls, bundle size, server-side performance, client data fetching, re-renders and rendering.