code-review-tool

Review your coding agent's diff in a terminal pane and send line comments back to Claude Code, Codex, OpenCode or Pi. A herdr plugin.

OrcaCode Review — the open code review harness. Multi-model reviews, merge gates, and no markup. Pay only for inference.

Self-Hosted LLM Code Review Agent for GitLab, Forgejo and GitHub. Local LLM support

Evidence-based code review skills for Codex and Claude Code

Multi-agent code review for Claude Code: parallel review by Codex (GPT-5.5), Gemini 3.1 Pro, and 4 Claude specialist subagents (security, performance, logic, regression)

The agent-first IDE: run a fleet of coding agents (Claude Code, Codex, Gemini, Copilot, Antigravity) across isolated git worktrees, with a commit-time review gate. macOS · Windows · Linux.

Triage for AI-generated code review: grades each changed diff line green/yellow/orange by impact, so reviewers spend attention where it matters. High-signal, low-noise (0 false blocks), deterministic — in your coding agent (MCP), editor (VS Code), and CLI/CI.

🔁 double-check — force your AI to review its own work and catch the bugs & hallucinations it missed the first time. A Claude Code self-review skill. 让 AI 复盘自审,揪出它漏掉的 bug 和幻觉

Local, private pre-push AI code review — no packaging needed

SystemGroup's Code Review Agent: Smith began as an Agent, an AI program in the Matrix programmed to keep order within the system by terminating human simulacra that would bring instability to the simulated reality, as well as any rogue programs that no longer serve a purpose.

Autonomous AI-powered GitHub PR reviewer built with LangGraph and TypeScript. Features parallel LLM-based code analysis, hallucination filtering, idempotent review publishing, webhook handling, and automated evaluation using precision, recall, and F1.

🐇 CodeRabbit Desktop Client for Windows – automated AI-powered code review with inline suggestions, PR analysis, and team collaboration tools. Native app for Windows 10 and 11 (2026). One‑click installer.

The human review step for AI-written changes: your agent proposes, you decide — inline diff in the browser, line comments, per-chunk apply, verdict back to the agent as JSON. Ships as an agent skill.

Panel Review: a guardian agent for AI coding decisions. Four frontier models review an agent's riskiest diffs and commits, with local review gates it cannot silently skip. Install: npx @truverifai/init

Desktop app for git worktrees: browse files, review diffs with line-anchored notes, and keep a terminal per checkout to run any coding agent. macOS and Linux.

Review your coding agent's changes like a pull request — locally, with inline comments and threaded follow-ups, then hand the marked-up review back to the agent

Agentic AI code review for GitHub Pull Requests via opencode. Structured, project-aware, remembers previous passes. Works free, no API key required.

AI code reviewer for Github PRs that understands your entire codebase, not just the diff. Debates every finding across multiple agents to kill false positives, and writes clean 1-click fixes right in your PR.

A Stop hook that won't let the turn end until fresh-context reviewers have read the diff, their findings are verified, and the loop converges — or honestly says it didn't.

Universal Expert Skill Suite & Dual-Scope Memory Engine for AI Coding Agents (Cursor, Claude Code, Windsurf, Copilot, Antigravity)