skill

AI4Math-Paper-Reading

AI4Math skills for structured math-paper reading, theorem dependency analysis, and paper-to-skill extraction.

VeryMath/AI4Math-Paper-Readingcommunitylistedunreviewed

Set up

Skills are installed by placing their folder where your agent loads skills.

Download ZIP from GitHub Open the repository

The download comes from github.com/VeryMath/AI4Math-Paper-Reading itself, not from our server, so what you get is exactly what its author published.

Install instructions come from the README

Skills do not use an MCP config. Follow the steps in the README below.

README

synced Oct 21, 2026

AI4Math · Paper Reading

Structured workflows for reading mathematical papers, extracting proof dependencies, and turning papers into reusable AI4Math skills.

中文说明 · Contributors · Skill packages · Installation · Quick start · Security model

What This Repository Is

This repository is the AI4Math home for paper-reading skills. It does not try to be a single monolithic agent. Instead, each package under skills/ owns a focused workflow, its own README, and the agent instructions needed to run it.

Use the root page as the public map, then open the package that matches your task.

Skill Packages

Package	Use it for	Start here
`paper-to-skill`	Run the full paper-to-skill workflow: PDF preparation, paper triage, proof-pattern extraction, and cross-paper synthesis.	`README` · `SKILL`
`math-paper-reading`	Read papers deeply, extract theorem dependencies, build proof pathways, and manage local reference notes.	`README` · `SKILL` · `router`
`after-ocr`	Audit and repair formula-heavy OCR Markdown with continuous coverage, evidence logs, and multi-pass reconciliation.	`README` · `SKILL`
`graph-theory-paper-reading`	Deep-read graph theory papers and produce structured LaTeX reports with proof-tree artifacts.	`README` · `SKILL`

Installation

The recommended path is AI-assisted installation: ask your coding agent to clone or update this repository, read the Skill instructions, install the entrypoints, and verify discovery.

Please install these AI4Math Skills for me.

Repository: https://github.com/VeryMath/AI4Math-Paper-Reading.git
Branch: main
Skill paths:
- skills/paper-to-skill
- skills/math-paper-reading
- skills/after-ocr
- skills/graph-theory-paper-reading

Steps:
1. Clone or update the repository locally.
2. Read README.md, SKILL.md, AGENTS.md if present, and each target Skill entrypoint.
3. If this environment supports local Skill discovery, link each directory that contains SKILL.md into the local skills directory.
4. Keep shared sibling support directories in place when a Skill depends on them.
5. Verify that the installed Skills are discoverable.
6. Tell me the installed paths, whether a restart is needed, and give me one test prompt.

Manual fallback for Codex-style local discovery:

git clone https://github.com/VeryMath/AI4Math-Paper-Reading.git
cd AI4Math-Paper-Reading
mkdir -p ~/.codex/skills
ln -s "$PWD/skills/paper-to-skill" ~/.codex/skills/paper-to-skill
ln -s "$PWD/skills/math-paper-reading" ~/.codex/skills/math-paper-reading
ln -s "$PWD/skills/after-ocr" ~/.codex/skills/after-ocr
ln -s "$PWD/skills/graph-theory-paper-reading" ~/.codex/skills/graph-theory-paper-reading

If your agent uses a different local Skill directory, replace ~/.codex/skills with that configured path.

Quick Start

Clone the repository and choose a package:

git clone https://github.com/VeryMath/AI4Math-Paper-Reading.git
cd AI4Math-Paper-Reading

For reusable skill extraction, start with:

skills/paper-to-skill/SKILL.md

For structured paper reading, start with:

skills/math-paper-reading/agent_router.md

For OCR cleanup or graph-theory-specific reading, start with:

skills/after-ocr/SKILL.md
skills/graph-theory-paper-reading/SKILL.md

Repository Layout

AI4Math-Paper-Reading/
├── README.md
├── README.zh-CN.md
├── SKILL.md
└── skills/
    ├── paper-to-skill/
    ├── math-paper-reading/
    ├── after-ocr/
    └── graph-theory-paper-reading/

Package-local examples are illustrative fixtures. Derived outputs from private papers should stay outside git unless they are intentionally sanitized examples.

Validation

Run the repository tests and any changed package's own checks:

python3 -m unittest discover -s tests -v
python3 -m unittest discover -s skills/after-ocr/tests -v

If you use Codex's local skill validator, run it against every changed package directory.

Security and Scope

Do not commit private papers, unpublished notes, local databases with personal annotations, API keys, .env files, or generated caches. Keep public examples small, source-attributed, and safe to redistribute.