Find plugins

by Alvis

General code writing tools including quality checks, testing, architecture, and implementation support

by XHXIAIEIN

Look up, read, edit, check and generate Construct 3 folder projects against the bilingual Construct3-RAG schemas.

by Dash0

OpenTelemetry observability for OpenAI Codex sessions. Captures prompts, tool calls, MCP calls, sub-agent activity, and token usage as OTel traces.

by baleen37

Datadog observability guidance via the pup CLI

by Stellars Henson

Data science project standards, notebook scaffolding, rich output styling, project creation via copier, and compliance review.

by Bentoware

Deckhand connects your AI assistant to your live Anki collection so it can see your decks, turn your sources into well-made cited cards, and help you fix weak ones — with you approving every change.

by lazygophers

深度研究插件（2026架构）- Task-Centric任务驱动型研究系统。DGoT动态优化降低50%成本，Agentic RAG精度99%，支持8种输出格式

development-discipline is an open-source Codex plugin indexed from GitHub with 12 skills. Review its manifest and source before installing.

by John Wilger

John's consolidated development workflow, engineering standards, worktree, task, review, and agentic-system tooling.

by Stellars Henson

Critical document analysis with persona-driven risk scoring, iterative scorecard evaluation, and versioned corrections. Includes adversarial-review - hostile red-team review of code and artefacts via fresh claude -p reviewers with eleven pluggable expert adversaries.

by Ray Lee

Grade and converge a repo's documentation system as an AI-agent context source — six-dimension star rating (completeness, correctness, freshness, linkage, consistency, economy) including entry-file token cost, with an improvement loop that fixes docs until target ratings are met.

by Stellars Henson

Structured document processing: source grounding (lexical + semantic + NLI entailment), Bayesian calibration, uniformization, PDF manipulation, and validation. Process input documents into verified, quality-controlled outputs.

Local, human-in-the-loop message bus for coordinating parallel AI coding-agent sessions on one machine

by get-dre

An interactive guide to DRE, the Declarative Reporting Engine: install it, connect a source, write a report, run it and deliver it.

by liuyuchen

给 ego lite 的可插拔决策层：默认 System One（Jev），可换本地 / 其他 OpenAI 兼容后端；执行层带 fail-closed 护栏。

engineering-standards is an open-source Codex plugin indexed from GitHub with 2 skills. Review its manifest and source before installing.

by Alvis

Documentation creation, code design, product strategy, and Notion integration for knowledge management

eval-case-reporter is an open-source Codex plugin indexed from GitHub with 1 skill. Review its manifest and source before installing.

by sunhome243

Compact Figma MCP with validated batch ops, multi-channel routing, library automation, and design skills.

by gyuha

Agent-engineering workflow skills for Claude Code, Codex, and opencode — twenty-two fg-* skills: the four-stage loop (fg-ask → fg-run → fg-learn → fg-done) plus outside-the-loop utilities, including fg-security (vendored security audit) and fg-debug (vendored diagnosis only: active failures return to fg-run; independent diagnoses route by size, durable eval, and all-exit safe cleanup).

by Gregory Block

The research debate engine for Special Circumstances: an additive blue side builds, a subtractive red side audits and owns the pass/fail gate, an engine runs the mechanics and the bench rules the final compromise. Best-of-N, the Catechism (a worth-our-time decision), and the full record preserved.

by Alvis

Tools for creating and managing Claude Code, Codex, and Grok Build configuration files

by JadeYingWah

Delivery-discipline behavior overlay for AI agents: a pre-implementation gate, resource survey, evidence discipline, and hands-on acceptance. Model-agnostic; it disciplines process, it does not raise the reasoning ceiling.

by Gregory Block

Trajectory evidence for Special Circumstances: what a session actually did, as against what it reported. Captures each seat's own transcript by path at SubagentStop, and reads it back with provenance on every row — so an inspection cites the primary record rather than a summary of it.