Find plugins

Evaluate a codebase with a structured engineering scorecard.

by Becos

Incrementally ships each Claude Code turn's prompt and text reply to the tenant's gbrain, together with the commands run, tool output, code edits, and the paths of files edited, and the repository, branch, author and commit the turn happened in. Bodies are captured by default and are secret-redacted, withheld entirely for sensitive paths, and capped before they are stored; each category has its own opt-out (VONIC_CAPTURE_COMMANDS/OUTPUTS/CODE/FILES=0). Thinking is never sent unless you opt in (VONIC_CAPTURE_THINKING=1, off by default). After installing, run /codecollab:login, then /codecollab:backfill for existing sessions.

by Full Stack Skills / PartMe.AI

Cross-language code lint enforcement for AI coding assistants (ZCode, Claude Code, Codex CLI, Kimi Code): Java, Rust, TypeScript, Python. PostToolUse hook auto-runs the native linter on every AI-written file and blocks on failure in strict mode.

by Full Stack Skills / PartMe.AI

Cross-language code lint enforcement for AI coding assistants (ZCode, Claude Code, Codex CLI, Kimi Code): Java, Rust, TypeScript, Python. PostToolUse hook auto-runs the native linter on every AI-written file and blocks on failure in strict mode.

by Google LLC

Enterprise autonomous security auditing, zero-false-positive PoC verification, and context-aware patch remediation using Google Cloud CodeMender (cm) on Gemini Enterprise Agent Platform.

by codeonym

Design skills for Claude Code with a reproducible, headless toolchain: graphic design, illustration, photo editing, UI design, typography, print layout, 3D with Blender, video editing and screen recording. Every script runs in a versioned container (Blender, GIMP, Inkscape, Krita, Scribus, darktable, ffmpeg, ImageMagick, fonts…) so it works the same on Linux, macOS and Windows (WSL2).

by Alex

A role-shaped subagent fleet: eleven agents, the skills they preload, the hooks that keep the board honest, the board itself, and the workflows that organise multi-phase work.

by CodeRabbit AI

AI-powered code review in Codex, powered by CodeRabbit.

by CodeRabbit AI

Review local code changes with CodeRabbit's official CLI and safely triage or apply unresolved CodeRabbit GitHub PR feedback through the official code-review and autofix skills.

by blockchainian

Deliver a planned feature as parallel codex workstreams off Claude's critical path — implement: workstreams.txt from plan.md, worktree-pool execution with per-workstream checks and bounded retries, merge onto the session branch with restore-on-red, push, PR; review: a schema-structured local codex review of the delta into review.json.

by Hamza Aziz

Ask OpenAI Codex (ChatGPT) and Claude the same question through your own signed-in CLIs. They answer independently, critique each other, reply to the critiques, and agree on one answer.

by Becos

Capture each turn to your becos brain and recall relevant memory before every prompt.

by xelth.com

Runs `codex exec` as a read-only reviewer from Claude Code and records each consultation as files (brief, verbatim reply, JSON ledger, findings tracked by id), with per-reviewer thread lineages, a fail-closed provider preflight, a reviewer roster, a review panel and a second engine (Gemini through Google's Antigravity CLI agy); needs PowerShell 5.1 or 7, git, and the Codex CLI signed in to ChatGPT or given a [model_providers] table whose env key the user sets (agy optional, signed in by the user).

by ayooleo

Local Codex token usage reports and a live terminal footer.

by Sidkik

Invoke Codex agents with pinned models as Claude Code subagents, via the official codex plugin's shared companion runtime.

by Cameron Bennion

Four-phase plan hardening for Codex: DeepSeek-V4.1-Flash adversarially reviews the plan, then inspects the final code. Codex builds; DeepSeek grades.

by HernandoR

Offline installer for the DeepSeek V4 Flash custom subagent for Codex

by youzooyou

Run a Codex CLI review as a fresh, one-shot process (generic codex exec with a hand-built prompt) instead of a shared long-lived app-server session. Bundles a standalone /codex-review command.

A stronger default for Codex CLI: six independently-invocable workflow stages, specialist role cards, and an evidence ledger that makes 'done' auditable.

by teladi

The Hive plugin for controlling a sleeping Agentinnen fleet through a data-sparse local MCP server. codex-master-mcp and Masterjet remain compatibility aliases.

by menaje

A policy-enforcing Streamable HTTP MCP bridge from ChatGPT to local Codex.

by A1mAssist

Lights the Work Louder Codex Micro agent keys from Codex CLI hook events.

by Melodic Software

Coordinate Codex work and identify evidence-backed opportunities for reusable skills.

by Marc Meixner (Shredi)

Token-frugal multi-agent orchestration for Codex CLI: the same Requirements Ledger, guard hooks and tier-routed chair/worker discipline as the Claude Code orchestrator plugin, vendored from a shared core and adapted to Codex's native hooks/agents/skills.