AI Visuals
OpenAI Plugins
The OpenAI plugins repository contains Codex plugin examples: manifests and optional skills, app connections, MCP servers, agents, commands, hooks and assets.
Use it as a developer reference for packaging a Codex extension. The repository describes plugin examples, rather than the former ChatGPT plugin system.
What it is
Examples of Codex plugin bundles
Each example lives in plugins/<name>/ and requires a .codex-plugin/plugin.json manifest. Companion folders and configuration files add the capabilities that example uses.
Why it stands out
The bundle can combine several surfaces
A plugin can package skills alongside app or MCP connections, commands and other supporting pieces. The required manifest and optional pieces show how a small extension can grow without starting with every integration.
Availability
Public repository and marketplaces
The README points to a default marketplace and a separate marketplace for API-key login users. Inspect the example files and access setup before using a bundle in your own Codex environment.
Why it matters
What makes it useful
If a recurring Codex task needs both instructions and an external tool, these examples show how to package those pieces together. A builder can inspect the bundle rather than distributing its instructions and connection settings separately.
What to know
Where it fits
This sits in extension development and configuration. For visual work, the Figma example covers creating or updating screens, design libraries and Figma or FigJam files; Remotion covers programmatic videos and animations with audio, captions or 3D scenes. These are plugin workflows for Codex, with their own tools and connections. The repository does not replace Codex itself, authenticate an external service for you, or supply a standalone assistant.
Notable points
What stands out
The default marketplace is .agents/plugins/marketplace.json; API-key login users have .agents/plugins/api_marketplace.json. Those are separate catalog paths, so an example's presence in one is not enough to assume it appears in the other.
Before using
What to review
Start with the files in the specific example you want to adapt, including its manifest and any connection configuration.
Check the current Codex setup and login path against the repository's marketplace instructions.
Review the instructions, executable hooks and tool connections a bundle includes before granting access to private files or external accounts.
Check the original listing's documentation for current requirements, limitations, licensing and any relevant privacy or security considerations.
Reader fit
Who may find it relevant
Builders creating or adapting Codex plugin bundles.
Teams inspecting how reusable instructions and tool integrations are packaged.
Readers seeking a ready-to-use chatbot or information about historical ChatGPT plugins should use a different starting point.
Editorial note
Why LifeHubber lists it
The richer examples cover different jobs: Figma combines design-system material with tool connections, while the iOS bundle centers SwiftUI work. Choose an example by the task and companion files you need, rather than copying the largest bundle as a default.
Source links
Source materials
Reader note
About our listings
LifeHubber’s listings help you explore what’s available in AI and are not endorsements. We don’t independently verify every detail or guarantee that each listing is accurate, safe, actively maintained, or right for your needs.
What to explore next
Inspect the skills inside a plugin.
A plugin's manifest describes a bundle, but its skill instructions and dependencies deserve their own inspection before you connect the bundle to work.
More in Ecosystem
Keep browsing this category
Explore more AI ecosystem resources.
Docling
docling-project/docling
A document conversion toolkit with structured text, tables and layout, native chunking for retrieval, Markdown and JSON exports, and local model configuration.
gputaskman
shanselman/gputaskman
Terminal monitor for local GPU activity and per-process memory, with before and after snapshots and separate model buffer estimates where supported.
MiniMind
jingyaogong/minimind
A native-PyTorch learning project for training tiny language models from scratch, with public data and checkpoints, tokenizer, pretraining and alignment stages, LoRA, tool-use and agentic-RL examples, plus local inference paths.
For project maintainers
Listed here? You can use the badge.
If you maintain a project with a current LifeHubber listing, you may add the optional “Listed on LifeHubber AI Resources” badge to its README, docs, or website. No introduction or permission request is needed.
MORE ON LIFEHUBBER
Use AI on your own terms.
Explore more AI projects, stories, guides, tools and community insights across LifeHubber.