Skip to content
Back to search

latex-paper-skills

A modular skill-based framework for writing, revising, and managing LaTeX academic papers with AI assistance.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/yunshenwuchuxun--latex-paper-skills/badge.svg)](/listings/yunshenwuchuxun--latex-paper-skills)

HTML

<a href="/listings/yunshenwuchuxun--latex-paper-skills"><img src="/api/listings/yunshenwuchuxun--latex-paper-skills/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Prerequisites

- Python 3.8+ - LaTeX environment (`pdflatex` + `bibtex`, or `latexmk`) - AI agent runtime that reads `SKILL.md` (Codex CLI, Claude Code, etc.)

Full pipeline (topic → PDF)

Use the paper-from-zero skill. My topic is: <your topic>

Direct review paper

Use the arxiv-paper-writer skill. Write a review article about <topic>.

Direct empirical paper

Use the empirical-paper-writer skill. Write an experimental paper about <topic>.

Prompt templates: specify datasets and cloud execution

You can specify datasets, runtime budget, and an intended cloud platform directly in natural language.

Use the empirical-paper-writer skill.
Topic: <topic>.
Mandatory datasets: <Dataset A>, <Dataset B>.
Primary dataset: <Dataset A>.
Do not use private data.
Design a compute-aware experiment plan for a single A100 80GB.
Use the empirical-paper-writer skill.
Topic: <topic>.
Dataset: <dataset name> from local path <path>.
Target runtime: cloud A100 x1, max 8 hours.
Need a local smoke run first, then a full cloud run.
Keep the claims bounded by this compute budget.
Use the empirical-paper-writer skill.
Topic: <topic>.
Mandatory datasets: <dataset names>.
Target platform: AutoDL / Lambda / Slurm cluster.
Generate experiments/, configs/default.yaml, and an experiments README for cloud execution.
Assume I will run the jobs myself and then use results-backfill after real CSVs are available.

Notes: - The skill can design for your dataset and cloud budget. - The skill does not automatically provision or submit cloud jobs for you. - Real results still need to be produced outside the AI session and written to `paper/results/`.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 326 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # latex-paper-skills
  2. New: [Beginner Workflow Guide (Chinese)](BEGINNER_WORKFLOW.zh-CN.md)
  3. English | [简体中文](README.zh-CN.md)
  4. > Portable **AI agent skill bundle** for writing ML/AI academic papers —
  5. > from topic to compiled PDF, with verified BibTeX citations, gated
  6. > workflows, and multi-agent collaboration.
  7. ## What This Repo Is
  8. This repository is not a single paper project. It is a reusable skill bundle for AI-assisted academic writing.
  9. In this public snapshot, the repo keeps:
  10. - the reusable writing skills under `.codex/skills/`
  11. - the repo-level workflow and documentation
  12. - two curated showcase projects: one empirical and one review
  13. The repo intentionally does **not** keep bulky local datasets, machine-local settings, or cache databases. For the empirical showcase, the raw dataset used during development has been removed, but the retained code, result files, figures, and final PDF still show that the workflow completed a real data-and-experiment loop.
  14. ## Status & Tested Models
  15. The showcases in this repo were developed and tested with **GPT-5.2 xhigh**. Using **GPT-5.4** is expected to yield better results. Test coverage is still limited — the workflow has significant untapped potential, though undiscovered bugs may exist.
  16. ## How It Works
  17. <p align="center">
  18. <img src="picture/pipeline-en.svg" alt="Pipeline Overview" />
  19. </p>
  20. **Optional multi-agent collaboration:**
  21. | Agent | Role | Tool |
  22. |-------|------|------|
  23. | Gemini | Breadth — literature expansion, keyword clusters, alternative framings | `gemini_bridge.py` |
  24. | Claude | Depth — claim stress-testing, evidence audit, routing judgment | `claude_bridge.py` |
  25. | Health check | Verify both CLIs are installed and API-reachable | `check-collaborators` |
  26. ## Skill Catalog
  27. The eight source skills live under `.codex/skills/`. Each has a `SKILL.md` (the executable spec an agent follows), plus optional `scripts/`, `assets/`, and `references/`. The installer exposes one `latex-paper-skills` entry containing all eight workflows and their resources.
  28. ### Core Pipeline
  29. | Skill | What it does |
  30. |-------|-------------|
  31. | **paper-from-zero** | Router. Topic → literature search → innovation framing → contribution map → evidence matrix → route to writer. |
  32. | **arxiv-paper-writer** | Review/survey executor. Gated IEEEtran LaTeX workflow with issues CSV contract, per-issue writing loop, citation verification, and QA. |
  33. | **empirical-paper-writer** | Experimental paper executor. Extends the review workflow with experiment matrices, result status tracking (`planned`/`placeholder`/`verified`), and evidence-claim mapping. |
  34. | **latex-rhythm-refiner** | Prose polisher. Varies sentence/paragraph rhythm, removes filler, strictly preserves all `\cite{}` positions. |
  35. | **results-backfill** | Back-fills real experiment results into an existing draft. Resolves placeholders, upgrades hypotheses to factual claims, generates figures. |
  36. ### Collaboration Layer
  37. | Skill | What it does |
  38. |-------|-------------|
  39. | **collaborating-with-gemini** | Breadth co-pilot via Gemini CLI. Structured JSON + session persistence. |
  40. | **collaborating-with-claude** | Depth co-pilot via Claude Code CLI. Claim stress-testing and evidence audit. |
  41. | **check-collaborators** | Health check — verifies CLI installation, auth, and API reachability. |
  42. ## Gated Workflow
  43. Both writer skills enforce strict gates:
  44. <p align="center">
  45. <img src="picture/gated-workflow-en.svg" alt="Gated Workflow" />
  46. </p>
  47. ## Quick Start
  48. ### Install from npm (recommended)
  49. The package is published on [npm](https://www.npmjs.com/package/latex-paper-skills)
  50. as **`latex-paper-skills@0.1.0`**, owned by `yunshenwuchuxun`. With **Node.js
  51. 22.20.0 or newer**, run this from your paper project:
  52. ```bash
  53. npx latex-paper-skills@latest --agent codex --copy --yes
  54. ```
  55. This installs the complete bundle for Codex in the current project. Replace
  56. `--agent codex` with `--agent claude-code` to use Claude Code. Then ask:
  57. `Use latex-paper-skills. My topic is: <your topic>`.
  58. Users do not need an npm account.
  59. To pin the published version or inspect the package:
  60. ```bash
  61. npx latex-paper-skills@0.1.0 --agent codex --copy --yes
  62. npx latex-paper-skills@latest --version
  63. npx latex-paper-skills@latest list
  64. ```
  65. For an optional user-level installation, add `--global` explicitly:
  66. ```bash
  67. npx latex-paper-skills@latest --agent codex --copy --yes --global
  68. ```
  69. The npm command uses the [Skills CLI](https://github.com/vercel-labs/skills)
  70. with the complete bundle already included in the npm package. It preserves all
  71. shared dependencies and has no install lifecycle hooks.
  72. ### Install from GitHub with the Skills CLI
  73. The existing GitHub installation remains available:
  74. ```bash
  75. npx skills@latest add yunshenwuchuxun/latex-paper-skills --skill latex-paper-skills --agent codex --copy --yes
  76. ```
  77. The installable directory is `skills/latex-paper-skills/`. It includes all eight
  78. workflows, `_shared`, scripts, and templates. Install this complete package;
  79. installing individual source folders under `.codex/skills/` omits their shared
  80. dependencies.
  81. ### Published packages and downloads
  82. - [ClawHub bundle](https://clawhub.ai/yunshenwuchuxun/plugins/latex-paper-skills):
  83. version **0.1.0**, install with
  84. `openclaw plugins install clawhub:@yunshenwuchuxun/latex-paper-skills@0.1.0`.
  85. - [LobeHub skill](https://lobehub.com/skills/yunshenwuchuxun-latex-paper-skills):
  86. open the listing for its installation or bundle download options. The
  87. marketplace listing version is **1.0.2**; its complete source bundle is
  88. **0.1.0**. These are separate version numbers.
  89. - [GitHub release v0.1.0](https://github.com/yunshenwuchuxun/latex-paper-skills/releases/tag/v0.1.0):
  90. [download the complete skill ZIP](https://github.com/yunshenwuchuxun/latex-paper-skills/releases/download/v0.1.0/latex-paper-skills-0.1.0.zip).
  91. All these distributions contain the complete suite. See
  92. [distribution and publishing](distribution/README.md) for format and license
  93. details.
  94. ### Prerequisites
  95. - Node.js 22.20.0+ for the npm installer
  96. - Python 3.10+ for the paper workflow scripts
  97. - LaTeX environment (`pdflatex` + `bibtex`, or `latexmk`) to compile PDFs
  98. - AI agent runtime that reads `SKILL.md` (Codex CLI, Claude Code, etc.)
  99. Node.js runs the installer. The paper workflows still need their Python and
  100. LaTeX environments, along with any dependencies required by your experiments.
  101. Optional collaboration also requires installed, authenticated Gemini and Claude
  102. CLIs.
  103. ### Full pipeline (topic → PDF)
  104. ```text
  105. Use the paper-from-zero skill. My topic is: <your topic>
  106. ```
  107. ### Direct review paper
  108. ```text
  109. Use the arxiv-paper-writer skill. Write a review article about <topic>.
  110. ```
  111. ### Direct empirical paper
  112. ```text
  113. Use the empirical-paper-writer skill. Write an experimental paper about <topic>.
  114. ```
  115. ### Prompt templates: specify datasets and cloud execution
  116. You can specify datasets, runtime budget, and an intended cloud platform directly in natural language.
  117. ```text
  118. Use the empirical-paper-writer skill.
  119. Topic: <topic>.
  120. Mandatory datasets: <Dataset A>, <Dataset B>.
  121. Primary dataset: <Dataset A>.
  122. Do not use private data.
  123. Design a compute-aware experiment plan for a single A100 80GB.
  124. ```
  125. ```text
  126. Use the empirical-paper-writer skill.
  127. Topic: <topic>.
  128. Dataset: <dataset name> from local path <path>.
  129. Target runtime: cloud A100 x1, max 8 hours.
  130. Need a local smoke run first, then a full cloud run.
  131. Keep the claims bounded by this compute budget.
  132. ```
  133. ```text
  134. Use the empirical-paper-writer skill.
  135. Topic: <topic>.
  136. Mandatory datasets: <dataset names>.
  137. Target platform: AutoDL / Lambda / Slurm cluster.
  138. Generate experiments/, configs/default.yaml, and an experiments README for cloud execution.
  139. Assume I will run the jobs myself and then use results-backfill after real CSVs are available.
  140. ```
  141. Notes:
  142. - The skill can design for your dataset and cloud budget.
  143. - The skill does not automatically provision or submit cloud jobs for you.
  144. - Real results still need to be produced outside the AI session and written to `paper/results/`.
  145. ## Shared Script Engine
  146. `.codex/skills/arxiv-paper-writer/scripts/` contains the core tooling shared by both writers:
  147. | Script | Purpose |
  148. |--------|---------|
  149. | `arxiv_registry.py` | arXiv metadata/BibTeX cache (SQLite) |
  150. | `compile_paper.py` | LaTeX compilation (latexmk or pdflatex+bibtex) |
  151. | `citation_policy.py` | Citation audit (bib/tex consistency, lint) |
  152. | `source_ranker.py` | Source quality scoring |
  153. | `style_profile.py` | Target venue style checking |
  154. | `issue_workflow.py` | Issue execution helpers |
  155. | `bootstrap_ieee_review_paper.py` | Scaffold IEEEtran project skeleton |
  156. | `create_paper_plan.py` | Generate paper plan from outline |
  157. | `validate_paper_issues.py` | Validate issues CSV integrity |
  158. Shared utilities (`paper_utils.py`, `source_policy_utils.py`) live in `.codex/skills/_shared/`.
  159. ## Non-Negotiable Rules
  160. - **No prose before approval** — `main.tex` stays skeleton-only until the plan is approved and the issues CSV exists.
  161. - **Issues CSV is the contract** — update status per issue; only mark `DONE` when acceptance criteria are met.
  162. - **Citations must be verified** — every citation is checked against an online source before entering `ref.bib`.
  163. - **Never fabricate** citations, results, or significance claims.
  164. ## Public Showcases
  165. This public snapshot keeps two curated showcase projects:
  166. | Project | Type | What it demonstrates | Start here |
  167. |--------|------|----------------------|------------|
  168. | `projects/rt-inflow-forecast-closed-loop` | Empirical paper | Topic framing, experiment design artifacts, local dataset integration, verified result files, figures, and back-filled paper writing | `README.md`, `paper/main.pdf`, `paper/results/`, `experiments/README.md` |
  169. | `projects/peft-survey-2022-2026` | Review paper | Topic framing, outline/plan approval, issues-driven writing, literature organization, citation-verified review drafting, and final paper | `README.md`, `main.pdf`, `plan/`, `issues/` |
  170. If you only want the fastest tour:
  171. 1. Open `projects/rt-inflow-forecast-closed-loop/README.md` for the empirical route.
  172. 2. Open `projects/peft-survey-2022-2026/README.md` for the review route.
  173. 3. Read `BEGINNER_WORKFLOW.zh-CN.md` if you want the end-to-end workflow explanation in Chinese.
  174. ## Project Structure
  175. ```
  176. latex-paper-skills/
  177. ├── .codex/skills/
  178. │ ├── paper-from-zero/ # Router: topic → writer skill
  179. │ ├── arxiv-paper-writer/ # Review paper executor + shared scripts
  180. │ ├── empirical-paper-writer/ # Empirical paper executor
  181. │ ├── results-backfill/ # Back-fill real results into draft
  182. │ ├── latex-rhythm-refiner/ # Prose polisher
  183. │ ├── collaborating-with-claude/ # Claude Code bridge
  184. │ ├── collaborating-with-gemini/ # Gemini CLI bridge
  185. │ ├── check-collaborators/ # CLI health check
  186. │ ├── _shared/ # Shared utilities across skills
  187. │ └── _orchestration/ # Workflow orchestration config
  188. ├── skills/latex-paper-skills/ # Complete installable bundle
  189. ├── bin/latex-paper-skills.mjs # npm installation command
  190. ├── package.json # npm metadata and installer dependency
  191. ├── distribution/ # Publishing guide and package metadata
  192. ├── scripts/build_distribution.py # Rebuild and verify distribution artifacts
  193. ├── projects/
  194. │ ├── rt-inflow-forecast-closed-loop/ # Empirical showcase
  195. │ └── peft-survey-2022-2026/ # Review showcase
  196. ├── picture/ # SVG diagrams for README
  197. ├── ARCHITECTURE.md # Detailed architecture analysis
  198. ├── BEGINNER_WORKFLOW.zh-CN.md # Beginner workflow guide (Chinese)
  199. ├── README.md
  200. └── README.zh-CN.md
  201. ```
  202. ## Distribution for Maintainers
  203. Source development stays under `.codex/skills/`. Before a future release, choose
  204. a new version and update the version fields described in the
  205. [publishing guide](distribution/README.md#standalone-npm-package). From the
  206. repository root, rebuild and verify the complete distribution, then prepare the
  207. npm tarball:
  208. ```bash
  209. npm ci
  210. npm run build:distribution
  211. npm run check:distribution
  212. npm test
  213. npm pack --pack-destination dist
  214. npm publish --dry-run --access public
  215. ```
  216. Review the package contents, authenticate as the intended publisher, and publish:
  217. ```bash
  218. npm login --auth-type=web
  219. npm whoami
  220. npm publish --access public
  221. ```
  222. Complete npm's browser 2FA verification when prompted. Version `0.1.0` is already
  223. published; subsequent releases require a new version. Commit the rebuilt bundle
  224. with its source changes so GitHub installers receive the same contents. See the
  225. [distribution and publishing guide](distribution/README.md) for the GitHub
  226. Release, ClawHub, and LobeHub workflows.
  227. ## Related
  228. - Forked from [renocrypt/latex-arxiv-SKILL](https://github.com/renocrypt/latex-arxiv-SKILL)
  229. - Workflow inspired by [appautomaton/agent-designer](https://github.com/appautomaton/agent-designer) (issue-driven development)
  230. - Uses [IEEEtran](https://ctan.org/pkg/ieeetran) LaTeX class
  231. ## Star History
  232. <p align="center">
  233. <a href="https://star-history.com/#yunshenwuchuxun/latex-paper-skills&Date">
  234. <picture>
  235. <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/image?repos=yunshenwuchuxun/latex-paper-skills&type=Date&theme=dark" />
  236. <source media="(prefers-color-scheme: light)" srcset="https://api.star-history.com/image?repos=yunshenwuchuxun/latex-paper-skills&type=Date" />
  237. <img alt="Star History Chart" src="https://api.star-history.com/image?repos=yunshenwuchuxun/latex-paper-skills&type=Date" />
  238. </picture>
  239. </a>
  240. </p>
  241. ## License
  242. This repository is released under the [MIT License](LICENSE).
  243. ## Community
  244. - [LinuxDO](https://linux.do/)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.