Havuz

Ghost personas review your code before it ships. Find edge cases, race conditions, and business logic bugs before production.

Local, live code review for the AI age - read what the agent wrote and send feedback straight back

Kimi for Claude Code, Codex, and DeepSeek Harness: code questions, independent reviews, bounded fixes, autonomous goals, parallel tasks, and

AI-powered code review agent for GitHub pull requests. Self-hosted GitHub App using PI.

Staff-engineer layer for AI coding agents: routes per-edit architectural constraints and requires verification. Not another AGENTS.md.

Kit de la clase Adversarial Review (Imperio Agéntico): app de demo con 43 fallas + prompts de auditoría UI/UX/seguridad + checklist + report

Semantic diffs and merges for UnityYAML assets.

Research-backed Harness Engineering study repo with a runnable Python MVP built around Plan -> Work -> Review.

Proof of Understanding. Starting with GitHub: show that you understand the change before it merges.

Runtime behavior diffs for pull requests

Self-hosted AI pull request review bot on Pi Durable. Comment /dureview to get P0–P3 inline findings and a check run, using your own ChatGPT

Claude Code delegates coding to isolated CLI agents it doesn't trust (Codex, OpenCode, Pi, Pythinker), freezes what they produce, verifies i

Pi 一站式扩展：并行子代理、自动修复的多模型审查、清爽界面 · All-in-one Pi extension: parallel sub-agents, self-fixing multi-model review, a calm UI.

Find vulnerabilities. Ship secure. — Senior security-engineer skill for AI coding agents. OWASP Top 10, CWE Top 25, secrets detection, IaC +

Opinionated Go skill for Claude Code, Codex, Cursor, Copilot & OpenCode — architecture, modern Go, testing, code review, MCP & agent pattern

Run AI coding agents unattended for hours and ship PRs worth merging. Cybernetics-based multi-agent orchestration + cross-LLM peer review fo

Your AI works the night shift — read-only reviews of your projects while you're busy, one digest every morning.

Claude Code skill that delegates implementation to GPT-6.1 (whatever is the latest) Sol via Codex CLI, Claude plans and reviews the diff, So

Diff filter that expands every hunk to include the enclosing function or block via tree-sitter. Pager for git, gh, or lazygit; viewer for co

Claude Code skill that quizzes you on the code Claude wrote — verify you understood it before shipping.

Open & nano reimplementation of Sakana Fugu. A repo-native, multi-agent coding loop powered by 9+ LLMs (isolated via Claude Code) and an ind

Generative UI for working with coding agents. Review, annotate, brainstorm, and give feedback through task-specific desktop workbenches.

Local git diff viewer

Fable-inspired verification-first workflow skill for Codex

Cybersecurity knowledge graph for code review with evidence-backed answers

Governed local runtime for AI coding agents: task lifecycle, mandatory gates, reviews, doc-impact checks, and auditable completion.

A plan-driven, multi-agent coding loop: separate planner, coder, and reviewer roles, each on the provider and model you choose, work togethe

MOVED to dsh-github-bot — https://github.com/PerryLink/dsh-github-bot. This repository is a pointer/backup; the old npm package @perrylink/d

Evidence-backed TS/JS repo structure lens for Claude Code

Your engineering crew for Claude Code — 12 specialists plan, build, review and ship in any stack; the rules that matter are enforced, not re

Quiet, risk-scaled engineering methods for Codex. Method as exponent, not machinery.

Claude Code plugin that stops "almost done" loops — declare your stages in workflow.json, runs them as a state machine that auto-retries unt

DeepSeek Harness plugin: 15 obra/superpowers engineering skills, bilingual descriptions, per-skill toggles | DeepSeek Harness 插件：15 个 obra/s

Logic-first AI code review via semi-formal execution tracing (Premises → Trace → Divergence → Trigger → Remedy). Catches behavioral bugs, ty

Agent-assisted vulnerability research across Rust, Android APKs, and AI agents: independent discovery, skeptical triage, verification, and p

把实现计划派发给另一个 AI 执行，你只负责审——纯 CLI 的协调者/executor 两角色协作工具（opencode / Claude Code / grok / codex，支持远程开发机）

Evidence-first multi-agent debate skill: get a second opinion by pitting Codex × Claude Code (or GLM/DeepSeek/Qwen) to independently review,

A gated agile delivery process for Claude Code: an AI product team (PO, Designer, EM, Reviewer, QA, Release) where every feature must pass a

inplan — a Markdown editor for human ⇄ coding-agent planning. CLI + desktop editor.

Local, deterministic review of Git changes: weakened tests and CI gates, security boundaries, blast radius. CLI, GitHub Action, MCP server,

🎵 Grade your vibe-coded project. One command, instant letter grade across security, quality, dependencies, and testing.

Tends your codebase while you're away. An always-on AI engineer that reviews your logs, implements your tickets and keeps your pull requests

Local two-agent collaboration for Claude Code, Codex, and Grok Build. Keep your native sessions and tools; relay, review, and audit in a sha

Codex plugin that lets Codex consult local Claude Code for advice, reviews, adversarial checks, and rescue tasks.

Chrome/Edge extension: stack-aware GitHub PR list + fast in-page review modal

The loop never sleeps — you finally can. Agents automate the entire development lifecycle (write, review, merge, learn) end to end, with a p

🧶 Your code-review inbox, in the macOS menu bar

Local-first repo maps for coding agents—ranked files, test routes, risks, CLI/MCP/GitHub Action, and public GitHub URLs.

Open-source repository standard, starter template, validator, and multi-agent workflow for making vibe-coded and AI-generated software under

No API keys required. Transform the native ChatGPT, Claude, and Gemini web interfaces into fully autonomous local coding agents.