Every plugin for OpenAI Codex on GitHub — repos with a .codex-plugin/plugin.json or a Codex marketplace. Install with codex plugin marketplace add owner/repo then codex plugin add name@marketplace. Many work in both tools; see Claude Code plugins.
Preset CLI skills for shell, scripting, and CI/CD workflows driven by the `sup` CLI (PyPI package `superset-sup`). Use only for CLI workflows; do not use for MCP-only work, and do not use as a substitute for direct HTTP/SDK code.
Analyze technical sources and produce Agent-engineering-positioned X content packs with evidence review, quality chain, reusable assets, and image assets.
Cria PRDs enxutos a partir de uma descrição de funcionalidade, com perguntas de clarificação, template padronizado e pesquisa web apenas quando necessária. Triggers — criar PRD, PRD, product requirements, requisitos de produto, especificação de produto, planejamento de feature.
Cria tasks incrementais a partir de PRD e Tech Spec, com aprovação high-level antes de escrever arquivos. Triggers — criar task, quebrar PRD em tarefas, plano de implementação, task planning, tasks.md, subtarefas.
TypeScript and Python SDK skills for building browser automation with Kernel's Typescript and Python SDKs. Use when writing code to control browsers programmatically.
Fully local visual plans and recaps — the agent writes markdown, a bundled zero-dependency server renders it in your browser with Mermaid diagrams, rich diffs, styled DB migrations, API call cards, and a read-only OpenAPI view, with live reload and a comment feedback loop. No remote services.
Keeps what the agent makes from reading as AI slop, in two skills. anti-slop-ui covers interfaces: no gradient-everything, rainbow palettes, pulsing badges for states that never change, fingernail-striped cards, emoji icons, default fonts, uppercase labels, scroll effects, screens with no hierarchy or no empty, loading, and error states, or landing-page hype on working screens. anti-slop-text covers prose: no invented details, inflated significance, promotional tone, weasel attributions, not-X-but-Y contrasts, manufactured rhythm, AI vocabulary, chat residue, restating conclusions, or PR text about what was left unchanged. Each checks its output against a catalog of known tells before it ships.
Hardlink-based skill sharing across projects — one source of truth per skill, hardlinks everywhere else. Puts the manage-skills CLI on PATH and ships its own skills.
Operate the bitdrift Capture platform with the bd CLI: investigate crashes, sessions, and app health, author workflows and dashboards, and set up Critical User Journey monitoring.
Fable 5 methodology as a Claude Code plugin: decompose by verification boundaries, verify against the world, choose next by plan-change. Ships fable-scout + fable-refuter agents.
Digest-bound lending-data releases with a resumable Ethereum USDC interval collector proven against live providers and attributing proxy logs by transaction position, unsigned evidence statements, reviewed credit views and address queries.
Generate crisp, perfectly smooth MP4 videos from web pages and animated visualizations by driving headless Chromium over CDP with deterministic frame-stepping, then encoding with ffmpeg. Use when making demo videos, explainer clips, or launch animations.