Code Review plugins

Compress Claude Code output by ~65% using caveman-style speech while preserving technical accuracy. Delegates code investigation, editing, and review to subagents that return compressed output. Generates ultra-compact commit messages and code reviews. Tracks token savings and session costs.

Automate PR review with specialized agents that analyze code quality, test coverage, error handling, comments, and type design, producing a prioritized action plan from git changes.

Orchestrate multi-agent teams for parallel code reviews across quality dimensions, hypothesis-driven debugging with competing root-cause investigations, and coordinated feature development with file ownership boundaries. Decompose tasks, design team compositions, and manage delegation with workload balancing.

Review code changes automatically by analyzing diffs against coding standards and spec/issue requirements using parallel sub-agents for thorough, multi-perspective analysis.

Automated multi-agent code review for GitHub pull requests with confidence-based scoring, checking for bugs, compliance with project guidelines, historical context, and past feedback.

Guide developers through receiving code review feedback with technical rigor: verify suggestions before implementing, avoid performative agreement, and push back with evidence-based reasoning when feedback is unclear or questionable.

Generate comprehensive documentation, code explanations, and tutorials from any codebase, with automated API docs, architecture diagrams, and step-by-step breakdowns. Includes AI-powered code review for security and performance insights.

Automates code refactoring, technical debt scanning, and legacy migration using the strangler fig pattern, with context restoration and AI-powered code review for security and performance.

Run multi-perspective code reviews across architecture, security, performance, and best practices using specialized agents, with automated PR description generation from git changes and optional framework-specific guidance.

Enforces a strict red-green-refactor TDD discipline with AI-assisted test generation, minimal-code implementation, and incremental refactoring. Includes autonomous code review agents that analyze pull requests for security, performance, and reliability.

Reviews code across correctness, readability, architecture, security, and performance before merging any change from any author.

Automate the full development lifecycle with AI agents for code review, research, design, and workflow orchestration. Includes browser automation, multi-agent code review, plan generation, documentation management, and Rails/Python/TypeScript code generation and auditing.

Automate pull request creation and review with conventional commits, quality gates, and AI-powered code analysis, while also generating new-hire onboarding playbooks.

Automates multi-agent AI code reviews for GitHub pull requests, analyzing security, performance, style, architecture, and accessibility across every PR.

Reviews TypeScript and JavaScript code changes against Metabase coding standards, catching style violations and quality issues during pull request or diff review.

Code intelligence powered by a knowledge graph for execution flow tracing, blast radius analysis, and augmented search across your codebase. Enables root cause debugging, safe refactoring, and PR review with dependency-aware impact analysis.

Generate self-contained HTML pages for visual explanations including diagrams, architecture overviews, diff reviews, implementation plans, slide decks, and project recaps, with automatic verification against source code and git history.

Review and resolve AI-powered code review comments on GitHub and GitLab pull requests directly from Claude Code using Greptile's API.

Review Clojure/ClojureScript code changes for compliance with Metabase coding standards, highlighting style violations and code quality issues during pull request reviews.

Reviews documentation changes (PRs, files, diffs) for compliance with the Metabase writing style guide, ensuring consistent docs in markdown.

Review and annotate markdown plans, code PR diffs, and assistant messages through an interactive UI, automatically integrating feedback into plan mode workflows.

Orchestrate multi-agent workflows with autonomous execution, cross-agent code review, plan merging, and structured visual recaps — reducing manual oversight on complex, codebase-heavy tasks.

Manage the full lifecycle of Claude Code skills: create, review against best practices, discover and install from registries, and enforce source-of-truth discipline across marketplaces and caches.

AI code reviews grounded in twelve classic engineering books — analyze PRs, architecture, tech debt, test quality, and codebase health with decay risk diagnostics, severity labels, and auto-fix capabilities.

Automate a structured GitHub bug-fix workflow: from issue creation, branch management, conventional commits, and auto-fixing PR feedback, through multi-perspective PR review and CI pipeline fix loops.

Automate comprehensive PR reviews by running multiple specialized agents that analyze git changes for code quality, test coverage, error handling, type design, comments, and simplification, then compile prioritized action plans.

Delegate coding tasks to the local OpenAI Codex CLI agent using a ChatGPT Pro OAuth flat-rate subscription, enabling code generation, refactoring, and review without per-token API charges.

Automatically review recent Git changes with actionable feedback on code quality, security, performance, testing, and documentation.

Reviews code changes by running git diff, inspecting modified files, and delivering prioritized feedback on quality, security, and maintainability.

Plan, implement, and review software tasks using structured .flow/ workflows with autonomous subagents, multi-model review gates, and codebase readiness assessments. Supports epic planning, task execution, code reviews, and project health checks.

Run an automated code review loop: Claude implements a task, Codex independently reviews the changes, and Claude iteratively addresses feedback until the review is approved.

Review code and receive structured feedback with line-level comments, a 1-10 quality rating, and prioritized action items covering performance, security, architecture, and testing.

Automate GitHub and Git workflows: create conventional commits, manage pull requests, review code, resolve PR comments, and keep branches clean. Includes secret scanning to prevent leaking credentials.

Analyzes review comments and change requests in a GitHub Pull Request, then automatically plans and applies fixes, runs tests, and updates the PR using GitHub CLI.

Reviews recent Git changes for code quality, security, performance, testing, and documentation, providing specific actionable feedback

Automatically fetches unresolved PR comments from the current branch and applies targeted code fixes to address reviewer feedback, streamlining the code review workflow.

Run git diff and inspect changed files to get prioritized feedback on code quality, security, and maintainability immediately after writing or modifying code.

Runs git diff after code changes, inspects modified files, and delivers prioritized feedback on code quality, security, and maintainability.

Review code changes and receive structured feedback with line-level comments, a quality rating, and prioritized action items covering performance, security, architecture, and testing.

Override the default submit_review MCP tool with a custom Bun-based workflow that enforces a 96-hour review timeout, enabling automatic plan review and approval routing.

Automatically review recent Git changes for code quality, security, performance, testing, and documentation, providing specific actionable feedback to improve code.

Automate code review resolution on GitHub PRs: analyze suggested changes and review comments, plan and apply fixes, run tests, and update the PR directly via the GitHub CLI.

Run structured, multi-domain code reviews with blast radius analysis, architecture audits, bug detection, test quality evaluation, and automated refactoring plans across Rust, shell scripts, Makefiles, and general codebases.

Review recent Git changes with automated analysis covering code quality, security, performance, testing, and documentation, delivering actionable feedback directly in the terminal.

Analyze code changes and receive structured reviews covering quality, performance, security, architecture, and testing, with line-level comments, a 1-10 rating, and prioritized action items.

Fetches unresolved comments from the current PR branch and applies targeted code improvements to resolve reviewer concerns, reducing manual back-and-forth.

Analyze GitHub PR review comments and suggested changes, then automatically plan, implement, test, and push fixes back to the PR using the GitHub CLI

Create reviewable HTML documents for design proposals and technical plans, serve live previews, and enforce a structured review workflow with threaded comments and blocking submit-review hooks.

Get second-opinion code reviews and bug analysis from ChatGPT Pro (web-only) directly within Claude Code by automatically packing relevant code with repomix and handling browser automation for login and submission.

Orchestrate solo development workflows: plan and execute PRs from task files, run automated review loops with external AI reviewers, manage backlog and git worktrees, generate handoff docs, and vet plans against best practices.

Define and enforce executable architecture fitness functions that validate code changes against quality guardrails, with automatic discovery of quality signals and generation of validation docs, triggered during code review via MCP.

Enforce rigorous code-review gates with technical feedback and subagent reviewer requests before allowing PR completion claims.

Orchestrates a multi-agent development pipeline where Claude plans and verifies while Codex CLI implements. Use commands for autonomous end-to-end implementation, cross-model code reviews, targeted fixes from review findings, and iterative improvement loops.

Fetch Qodo PR review issues for your current branch, apply fixes interactively or in batch, and reply to inline comments — catching issues before commit and enforcing organizational standards directly in your agent.

Review and analyze Claude Code session transcripts to improve prompting effectiveness, agent performance, and identify environment gaps. Export session conversations as Markdown with metadata and collapsible tool results. Analyze multiple sessions in parallel for cross-session patterns and actionable recommendations.

Automate OSS maintenance workflows: manage issues and PRs, perform code reviews, diagnose CI/CD pipelines, prepare releases, and analyze repository health for Python projects on GitHub.

Isolates subagents in parallel by provisioning separate git worktrees, preventing file conflicts, and merges completed work via cherry-pick or merge. Also enforces a custom code review workflow by intercepting review submissions and routing them through a CLI.

Instruments live production services with Lightrun to inspect variable values, execution durations, hit counts, and value distributions; guides deterministic debugging of errors with evidence-based diagnosis and PR-first fix proposals; and reviews pull requests with runtime production evidence including live samples and simulated patches.

Run a structured consensus protocol with isolated worker perspectives for high-risk development decisions, enabling autonomous resolution of GitHub issues and pull requests through goal definition, multi-angle analysis, implementation, and review.

Optimizes unstructured subagent prompts by scoring across 7 dimensions and restructuring with Anthropic best practices like XML tags and chain-of-thought, plus blocks MCP tool calls to enforce a custom code review workflow.

Automatically resolve GitHub PR review comments by analyzing suggestions, applying fixes, running tests, and updating the pull request via the GitHub CLI.

Verify your understanding of AI-generated code and plans through rubber duck questioning, with automatic hooks that remind you to commit after every Bash command and update specs after file edits, preventing rubber-stamping in AI-assisted workflows.

Provides a structured code review and improvement workflow: triages PR feedback and CI failures, reviews diffs across multiple quality dimensions, resolves merge conflicts with AST-aware merging, hardens tests, and implements approved specs through stale-safe code edits.

Collaborate on engineering RFCs: draft, review, and synthesize discussions, extract action items, compare alternatives, and recommend reviewers — all through chat without leaving Claude Code.

Automates PR creation, code review, and commit workflows via slash commands, with task decomposition into Linear issues, plan review, security hardening of GitHub Actions, and integrations with Notion, esa, and cloud auth.

Automatically review GitHub pull requests with multiple specialized agents that check for bugs, CLAUDE.md compliance, historical context, and prior PR feedback.

Runs a team of 5 specialist agents (backend, frontend, test, security, UX) to parallel-review git diffs and produce an integrated Markdown report for comprehensive code quality checks.

Generate comprehensive code documentation, architecture diagrams, and step-by-step tutorials from codebases, with automated CI/CD pipeline support and AI-powered code review for security and performance.

Run multiple AI agents in parallel to review plans and code, then synthesize their feedback and debate contradictions to produce a consensus verdict with automatic approval cycling.

Runs a structured multi-agent delivery pipeline: task planning, implementation, multi-reviewer code review, and layered verification with specialist agents for security, performance, reliability, and acceptance testing.

Deploy and manage full-stack web apps on Major.build from Claude — create, clone, start, and deploy projects with GitHub integration, manage databases and auth, and enforce a custom code review workflow via a Bun-based CLI.

Replaces Claude Code's built-in WebSearch and WebFetch with DuckDuckGo-powered alternatives, and enforces a custom code review workflow by intercepting the submit_review tool and routing it through a Bun-based CLI with extended timeout.

Run structured reconnaissance on unfamiliar codebases to surface hidden risks, log deviations from plans, quiz yourself on risky diffs before merging, validate UX with interactive prototypes, and verify comprehension of reference code before adapting it.

Automatically review and fix uncommitted git changes in an iterative loop using independent AI reviewer agents, continuing until quality score reaches 9.5/10 or no actionable comments remain.

A multi-skill agent suite for mobile app store review compliance, product/market analysis, code review, technical project management, end-to-end feature delivery, sales pitch roleplay coaching, and PRD drafting.

Provides a structured code review and verification pipeline that runs static analysis, linter checks, test execution, and visual UI inspection in parallel, producing a unified report with actionable findings. Also automates git workflows: creating pull requests with context-rich descriptions, summarizing branch changes, and generating release notes from commits or diffs.

Automate git and PR workflows with multi-stage code review, quality gate enforcement, and PR enhancement. Generate onboarding playbooks for new hires. Produce rich PR descriptions with change statistics and impact analysis.

Leverage OpenAI Codex as an adversarial reviewer for pull request diffs and superpowers specifications, engaging in a debate loop until Codex approves the changes or specs. Ensures correctness, security, reliability, simplification, and performance before merging or proceeding.

Develop and manage decentralized applications on the Freenet peer-to-peer network with AI-assisted workflows for environment setup, contract design, debugging, testing, code review, and release management.

Run structured multi-dimensional code reviews covering architecture, security, performance, and testing through phased subagent orchestration, plus auto-generate enriched PR descriptions from git changes

Orchestrate AI-assisted engineering workflows with a structured pipeline framework: TDD test-first cycles, parallel subagent task execution, QA acceptance pipelines, code review triage and request generation, isolated git worktree development, and housekeeping that syncs memory and indexes. Supports ambiguous-request routing and multi-skill dispatch.

Coordinate multi-agent workflows for GitHub project maintenance: triage issues, review PRs, implement changes, and manage cross-session work boards with automated testing and commit assembly.

Run autonomous development workflows while away: batch clarifying questions, hand off to an AFK agent that handles ambiguity, babysits PRs to green, and returns a structured summary. Enforce git conventions, multi-reviewer code gates, and mise-based reproducible dev environments.

Automate multi-faceted PR code review by launching specialized agents that analyze git changes for comments, tests, error handling, types, code quality, and refactoring, delivering a prioritized action plan.

Launch multiple specialized review agents against git changes to catch issues across comments, tests, error handling, types, code quality, and code simplification, then get a prioritized action plan before opening a PR

Perform semantic code research with ChunkHound: multi-hop search, architecture analysis, dependency mapping, and LLM synthesis. Route code reviews through a custom Bun-based CLI workflow.

Manage Linear issues, projects, teams, and notification inbox directly from Claude Code — create, update, query issues, filter and archive notifications, and enforce a custom code review workflow with configurable timeouts.

Orchestrates multiple specialized PR review agents against git changes to generate a prioritized action plan covering comments, tests, error handling, types, code quality, and simplification.

Enforces a custom code review workflow by intercepting submit_review MCP calls and routing them through a Bun-based CLI with extended timeouts

Automated code reviews on any code snippet, providing line-level feedback on quality, performance, security, architecture, and testing, along with a 1-10 rating and prioritized improvement suggestions.

Reviews code changes by running git diff and inspecting modified files, then provides prioritized feedback on quality, security, and maintainability.

Automatically analyze GitHub PR review comments and suggested changes, then plan and apply fixes, run tests, and update the PR using the GitHub CLI.

Launch multiple specialized review agents against git changes to catch issues across comments, tests, error handling, type design, code quality, and code simplification, then get a prioritized action plan before committing or opening a PR.

Automatically fetches unresolved GitHub PR comments for the current branch and applies fixes to address reviewer feedback, streamlining the code review workflow.

Automate the full development lifecycle for autonomous startups on workers.do: structured brainstorming, TDD with RED-GREEN-REFACTOR, and multi-agent code review to ensure quality.

Develop Go applications with expert guidance on concurrency, performance, and idiomatic patterns; scaffold production-ready projects, review code for issues, and generate comprehensive tests with automated agents.