Codex Academic Skills

English | 简体中文

A curated list of research-oriented skills that are usable in OpenAI Codex.

The goal is simple: if a skill is listed here, it should either be from the official Codex ecosystem or from a repository that explicitly supports Codex or the open Agent Skills workflow that Codex can use.

---

Table of Contents

• What Are Codex Skills?
• Inclusion Rules
• How To Use This List
• Skill List
• Installation and Usage
• License
• References

---

What Are Codex Skills?

Codex skills are folder-based instruction bundles that teach Codex how to handle a task more reliably.

A typical skill usually includes:

• a SKILL.md file with trigger rules and workflow guidance
• optional scripts, templates, and references
• a stable folder structure that Codex can discover from standard skill locations

In practice, a good skill works like a reusable research playbook: Codex loads it when needed, follows the instructions, and combines that guidance with your local repository context.

---

Inclusion Rules

This list only keeps skills that satisfy at least one of the following:

• official OpenAI Codex skills
• repositories that explicitly document Codex support
• repositories built around the open Agent Skills format that Codex can consume with minimal adaptation

This list intentionally removes:

• skills that are exclusive to other platforms, for example Claude Code-only skills
• document-oriented workflows that depend on platform-built-in capabilities and do not translate cleanly into reusable Codex skills
• repositories whose Codex compatibility is unclear

---

How To Use This List

This list works best as a research-workflow index. It helps narrow the search space first, then sends you back to the upstream SKILL.md for the actual details.

If you are new to the list, a quick task-based pass is usually enough:

• For research workflow design, task decomposition, and context management, start with sections 1 and 2.
• For paper writing, literature organization, and evidence summaries, start with sections 3 and 4.
• For demos, observability, data work, or experiment pipelines, start with sections 5 and 6.
• If a skill looks promising, open the upstream SKILL.md before deciding whether to install or rely on it.

---

Skill List

1. Planning and Workflow

Skill	What It Does	Link
project-development	Helps scope LLM projects, assess task-model fit, and design practical research-agent architectures.	Agent-Skills-for-Context-Engineering
notion-research-documentation	Research across Notion and turn scattered notes into briefs, comparisons, and reports with citations.	openai/skills
brainstorming-research-ideas	Guides structured ideation for discovering defensible, high-impact research directions.	AI-Research-SKILLs
creative-thinking-for-research	Applies cognitive-science creativity frameworks to help generate genuinely novel research ideas.	AI-Research-SKILLs

2. Deep Thinking and Research Framing

Skill	What It Does	Link
context-fundamentals	Explains how context works in agent systems and how to design high-signal task context.	Agent-Skills-for-Context-Engineering
context-degradation	Helps diagnose lost-in-the-middle, distraction, poisoning, and other context failure modes.	Agent-Skills-for-Context-Engineering
context-compression	Teaches how to summarize and compress long-running research or coding sessions without losing critical state.	Agent-Skills-for-Context-Engineering
advanced-evaluation	Covers LLM-as-a-judge workflows, rubric design, and bias-aware automated evaluation.	Agent-Skills-for-Context-Engineering

3. Writing and Scholarly Communication

Skill	What It Does	Link
doc	Codex-oriented DOCX workflow for creating or editing research reports with rendering and layout checks.	openai/skills
notion-research-documentation	Useful for writing research briefs, comparison notes, and structured evidence summaries.	openai/skills
hugging-face-paper-publisher	Publishes papers, links them to models or datasets, and generates professional research article pages.	huggingface/skills
ml-paper-writing	Writes publication-ready ML/AI/Systems papers with citation verification and venue-aware structure.	AI-Research-SKILLs

4. Literature Reading and Evidence Synthesis

Skill	What It Does	Link
notion-research-documentation	Synthesizes multi-source findings into literature summaries and cited research notes.	openai/skills
llamaindex	Builds document-ingestion and retrieval pipelines for paper corpora, notes, and private research archives.	AI-Research-SKILLs
faiss	Provides high-performance dense retrieval for large paper or note embedding collections.	AI-Research-SKILLs
sentence-transformers	Generates strong semantic embeddings for literature search, clustering, and retrieval.	AI-Research-SKILLs

5. Visualization and Presentation

Skill	What It Does	Link
gradio	Builds interactive demos and polished research interfaces for models, ablations, and prototypes.	huggingface/skills
hugging-face-trackio	Tracks and visualizes training metrics with real-time dashboards.	huggingface/skills
langsmith-observability	Adds tracing, evaluation, and monitoring for LLM apps and experiment pipelines.	AI-Research-SKILLs
phoenix-observability	Open-source observability for tracing, experiments, and real-time analysis of AI systems.	AI-Research-SKILLs
stable-diffusion-image-generation	Useful for generating figures, visual concepts, teaser graphics, or multimodal assets for presentations.	AI-Research-SKILLs

6. Data and Experimentation

This is an extra category I recommend keeping because modern research workflows depend on it heavily.

Skill	What It Does	Link
hugging-face-datasets	Creates, manages, queries, and transforms datasets on Hugging Face Hub.	huggingface/skills
hugging-face-evaluation	Adds and manages structured evaluation results for models and benchmarks.	huggingface/skills
hugging-face-model-trainer	Trains or fine-tunes LLMs with TRL on Hugging Face Jobs.	huggingface/skills
hugging-face-jobs	Runs compute jobs on Hugging Face infrastructure for evaluation, generation, or training workflows.	huggingface/skills
evaluating-llms-harness	Runs standardized academic benchmarks such as MMLU, HumanEval, GSM8K, and TruthfulQA.	AI-Research-SKILLs
serving-llms-vllm	Serves LLMs with high-throughput inference and OpenAI-compatible endpoints.	AI-Research-SKILLs

---

Installation and Usage

This repository is a curated list, not a unified marketplace. You usually install a skill from its original source repository into a Codex skill directory.

Install a skill in Codex

Common Codex skill locations include:

• repository scope: .codex/skills/<skill-name>/
• user scope: ~/.codex/skills/<skill-name>/

Example 1: install an official Codex-oriented skill from openai/skills

mkdir -p ~/.codex/skills
cd /tmp
git clone --depth 1 https://github.com/openai/skills.git
cp -R skills/skills/.curated/notion-research-documentation ~/.codex/skills/

Example 2: install a research skill from AI-Research-SKILLs

mkdir -p ~/.codex/skills
cd /tmp
git clone --depth 1 https://github.com/Orchestra-Research/AI-Research-SKILLs.git
cp -R AI-Research-SKILLs/20-ml-paper-writing ~/.codex/skills/ml-paper-writing

Use a skill in Codex

Once the folder is in a valid Codex skill location, you can invoke it naturally in your prompt.

Examples:

• Use the ml-paper-writing skill to turn this repo into a NeurIPS-style draft.
• Use brainstorming-research-ideas to generate three defensible project directions.
• Use notion-research-documentation to turn these source notes into a cited literature brief.
• Use evaluating-llms-harness to benchmark this checkpoint on MMLU and GSM8K.
• Use gradio to build a polished demo for this paper artifact.

Recommended usage pattern

1. Pick one skill for one clear bottleneck.
2. Start with a narrow task, not a giant workflow.
3. Read the upstream SKILL.md before relying on output.
4. For academic work, it is worth manually checking citations, claims, equations, dataset handling, and benchmark settings, especially when copyright, privacy, attribution, or research-integrity requirements matter.

---

License

The content of this repository is released under the MIT License.

Third-party skills linked from this list keep their own licenses. Always check the original repository before installing or redistributing anything.

If you notice a dead link, a compatibility change, or a clearly better entry for the list, a short issue or PR is enough.

---

References

• OpenAI Codex Skills
• openai/skills
• Orchestra-Research/AI-Research-SKILLs
• Agent-Skills-for-Context-Engineering
• huggingface/skills
• awesome-agent-skills
• Agent Skills Open Standard