Skip to content
Back to search

ai4s-skills

Open-source agent skills for AI for Science: topic exploration, literature survey, experiments, paper writing, and integrity audit — driven by any coding agent.

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/ai4s-research--ai4s-skills/badge.svg)](/listings/ai4s-research--ai4s-skills)

HTML

<a href="/listings/ai4s-research--ai4s-skills"><img src="/api/listings/ai4s-research--ai4s-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.

Run the installer from the project you want the skills in:

git clone https://github.com/ai4s-research/ai4s-skills

cd /path/to/your-project
/path/to/ai4s-skills/install.sh                              # all skills → ./.claude/skills
/path/to/ai4s-skills/install.sh paper-writer                 # or specific ones
SKILLS_DIR=~/.claude/skills /path/to/ai4s-skills/install.sh  # global instead

To install by hand, copy any `skills/<name>/` into `~/.claude/skills/` (global) or `<project>/.claude/skills/` (project).

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 · 215 lines

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

  1. <div align="center">
  2. [![AI4S Skills — agent skills for AI for Science](assets/banner.webp)](https://github.com/ai4s-research/ai4s-skills)
  3. Seven agent skills for AI-for-Science research — turn a research direction into
  4. literature surveys, runnable experiments, publication-grade papers, and integrity
  5. audits, with every citation, number, and figure traceable to its source.
  6. <p align="center"><b>English</b> · <a href="README.zh.md">中文</a></p>
  7. <p>
  8. <a href="LICENSE"><img src="https://img.shields.io/badge/License-MIT-blue.svg" alt="License: MIT"></a>
  9. <a href="https://doi.org/10.5281/zenodo.21297455"><img src="https://img.shields.io/badge/DOI-10.5281%2Fzenodo.21297455-1682D4" alt="DOI"></a>
  10. <img src="https://img.shields.io/badge/skills-7-success" alt="7 skills">
  11. <a href="http://makeapullrequest.com"><img src="https://img.shields.io/badge/PRs-welcome-brightgreen.svg" alt="PRs Welcome"></a>
  12. <a href="https://linux.do"><img src="https://img.shields.io/badge/Join-linux.do-orange" alt="linux.do"></a>
  13. </p>
  14. </div>
  15. ---
  16. ## Contents
  17. - [🧩 The skills](#the-skills)
  18. - [🔗 How they connect](#how-they-connect)
  19. - [✅ Authenticity](#authenticity)
  20. - [🔬 Example](#example)
  21. - [📦 Install](#install)
  22. - [🚀 Usage](#usage)
  23. - [🗂️ Repository layout](#repository-layout)
  24. - [🛠️ Included tools](#included-tools)
  25. - [🤝 Contributing](#contributing)
  26. - [⚖️ License](#license)
  27. - [🙏 Acknowledgments](#acknowledgments)
  28. ## The skills
  29. | Skill | Role | Primary output |
  30. |---|---|---|
  31. | [ai4s-agent](skills/ai4s-agent/SKILL.md) | Runs the four skills below in order | the full package |
  32. | [research-explorer](skills/research-explorer/SKILL.md) | Explore topics from a broad direction | `research_exploration.md`, `topic_matrix.md`, `literature_pre_survey.md` |
  33. | [literature-survey](skills/literature-survey/SKILL.md) | Write a literature survey | 6–20 pp PDF, 60+ real citations, LaTeX source, topic-specific figures |
  34. | [experiment-suite](skills/experiment-suite/SKILL.md) | Build an experiment package | design doc, runnable code, `results.json` with provenance, figures, report |
  35. | [paper-writer](skills/paper-writer/SKILL.md) | Write a research paper | 8–14 pp PDF, 200+ citations, 4–8 figures, tables |
  36. | [mindmap-render](skills/mindmap-render/SKILL.md) | Render a mindmap | image from a `topic_matrix.md` (Python script) |
  37. | [integrity-auditor](skills/integrity-auditor/SKILL.md) | Audit a paper's integrity | image / numerical / logical findings, 4-level evidence grading, `audit_report.md` |
  38. Each skill is a folder with a `SKILL.md` plus its own references, templates, and
  39. tools. MIT-licensed; works with Claude Code, Cursor, Codex, and Aider.
  40. ## How they connect
  41. ```
  42. direction
  43. │
  44. ▼
  45. [1] research-explorer ──▶ pick one $TOPIC
  46. │
  47. ├──▶ [2] literature-survey → survey PDF + bibliography.bib
  48. ├──▶ [3] experiment-suite → results.json + figures/
  49. └──▶ [4] paper-writer → paper PDF (reuses [2] and [3])
  50. integrity-auditor ──▶ audits any paper: external PDF / DOI / arXiv, or [4]'s output
  51. ```
  52. `ai4s-agent` runs steps 1–4 in order. Skills pass work to each other through a
  53. shared slug and the path `output/<skill>/<slug>/latest/`.
  54. ## Authenticity
  55. The focus of the project. Every skill enforces:
  56. | Principle | In practice |
  57. |---|---|
  58. | **Real citations** | Every BibTeX entry links to a URL the agent fetched in the same session; none from memory. |
  59. | **Labelled numbers** | Every number is marked `measured`, `simulated`, or `illustrative`; simulated values are never reported as measured. |
  60. | **Runnable experiments** | `experiment-suite` outputs runnable code and a `results.json` with provenance. Supply real results and they replace the simulated ones; the "simulated" disclosure is then removed. |
  61. | **Resumable runs** | Long tasks save progress after each step and continue from the last checkpoint, so a reported "done" reflects completed work. |
  62. | **Publication layout** | Readable booktabs tables; final-size raster QA; vector-PDF figures with embedded fonts, shared color-safe styling, and no generic diagram-tool output. |
  63. | **Review disclosure** | Every generated document states that domain-expert review is recommended. |
  64. | **Integrity checks** | `integrity-auditor` inspects a paper for image, numerical, and logical problems and grades the evidence. |
  65. ## Example
  66. A complete run from `experiment-suite` + `paper-writer`: **"Learning the Burgers
  67. Solution Operator with a Fourier Neural Operator"** — an 8-page paper backed by code
  68. the agent wrote and ran. Full artifact in [`examples/fno-burgers/`](examples/fno-burgers/)
  69. (paper, code, `results.json`, report).
  70. <div align="center">
  71. <table>
  72. <tr>
  73. <td><a href="examples/fno-burgers/paper.pdf"><img src="assets/paper/page-1.webp" width="230" alt="paper page 1"></a></td>
  74. <td><a href="examples/fno-burgers/paper.pdf"><img src="assets/paper/page-2.webp" width="230" alt="paper page 2"></a></td>
  75. <td><a href="examples/fno-burgers/paper.pdf"><img src="assets/paper/page-3.webp" width="230" alt="paper page 3"></a></td>
  76. </tr>
  77. <tr>
  78. <td><a href="examples/fno-burgers/paper.pdf"><img src="assets/paper/page-4.webp" width="230" alt="paper page 4"></a></td>
  79. <td><a href="examples/fno-burgers/paper.pdf"><img src="assets/paper/page-5.webp" width="230" alt="paper page 5"></a></td>
  80. <td><a href="examples/fno-burgers/paper.pdf"><img src="assets/paper/page-6.webp" width="230" alt="paper page 6"></a></td>
  81. </tr>
  82. </table>
  83. <sub><i>The 8-page paper (first 6 pages) — click any page for the full PDF.</i></sub>
  84. </div>
  85. - **Real code, really run** — `model.py` is a 1-D FNO; the full study runs in ~35 min on a laptop CPU.
  86. - **Measured results** — FNO 4.41% rel-L2 vs MLP 21.63% and CNN 67.69% (3 seeds); zero-shot super-resolution holds 4.2–4.7% from grid 128 to 1024.
  87. - **Real citations** — 22 references, each traceable to its source.
  88. Every number is `measured` (provenance in `results.json`); the paper states it was
  89. AI-generated and recommends domain-expert review.
  90. ## Install
  91. Run the installer from the project you want the skills in:
  92. ```bash
  93. git clone https://github.com/ai4s-research/ai4s-skills
  94. cd /path/to/your-project
  95. /path/to/ai4s-skills/install.sh # all skills → ./.claude/skills
  96. /path/to/ai4s-skills/install.sh paper-writer # or specific ones
  97. SKILLS_DIR=~/.claude/skills /path/to/ai4s-skills/install.sh # global instead
  98. ```
  99. To install by hand, copy any `skills/<name>/` into `~/.claude/skills/` (global) or
  100. `<project>/.claude/skills/` (project).
  101. ## Usage
  102. In Claude Code:
  103. > Use the literature-survey skill to write a survey on \<your topic\>.
  104. With Cursor, Codex, or Aider, point the agent at the skill file:
  105. ```
  106. Read skills/literature-survey/SKILL.md and its references/, then produce the survey
  107. for "<your topic>" as specified.
  108. ```
  109. Each `SKILL.md` directs the agent to read its `references/` first; those files hold
  110. the procedures for bibliography expansion, figures, layout, and quality checks.
  111. ## Repository layout
  112. ```
  113. ai4s-skills/
  114. ├── skills/
  115. │ ├── ai4s-agent/ SKILL.md + references/
  116. │ ├── research-explorer/ SKILL.md
  117. │ ├── literature-survey/ SKILL.md + references/ + templates/survey/
  118. │ ├── experiment-suite/ SKILL.md + references/ + figure_examples/
  119. │ ├── paper-writer/ SKILL.md + references/ + templates/paper/
  120. │ ├── mindmap-render/ SKILL.md + scripts/ + tests/
  121. │ └── integrity-auditor/ SKILL.md + references/ + forensics_tools/ + templates/ + tests/
  122. ├── tools/validate_skills.py structure / frontmatter validator (run in CI)
  123. ├── install.sh
  124. └── .github/workflows/ci.yml
  125. ```
  126. Each `SKILL.md` carries YAML frontmatter (`name`, `description`) so an agent can
  127. find and route to it.
  128. ## Included tools
  129. Small, single-purpose scripts the skills call. Each directory has its own
  130. `requirements.txt`.
  131. - `skills/integrity-auditor/forensics_tools/` — image duplication / ORB matching, panel splitting, channel checks, magnitude (Benford-style) consistency, decimal matching, spreadsheet aggregate consistency.
  132. - `skills/experiment-suite/figure_examples/` — a matplotlib style kit (`style_kit.py`) and worked figure examples.
  133. - `skills/mindmap-render/scripts/` — `generate_mindmap.py`.
  134. ## Contributing
  135. A new skill needs:
  136. 1. `skills/<name>/SKILL.md` with `name` and `description` frontmatter (`name` = folder name).
  137. 2. Optional `references/`, `templates/`, and tools.
  138. 3. No `import anthropic` / `import openai`.
  139. 4. `python tools/validate_skills.py` passing (CI runs it on every PR).
  140. See [CONTRIBUTING.md](CONTRIBUTING.md) and the [Code of Conduct](CODE_OF_CONDUCT.md).
  141. ## Citation
  142. If you use AI4S Skills in your research, please cite it:
  143. ```bibtex
  144. @software{ai4s_skills,
  145. author = {{The AI4S Skills Contributors}},
  146. title = {AI4S Skills: open-source agent skills for AI for Science},
  147. year = {2026},
  148. version = {0.1.0},
  149. doi = {10.5281/zenodo.21297455},
  150. url = {https://github.com/ai4s-research/ai4s-skills},
  151. license = {MIT}
  152. }
  153. ```
  154. The DOI above cites all versions; GitHub's **"Cite this repository"** button
  155. (generated from [`CITATION.cff`](./CITATION.cff)) provides the same reference
  156. in APA and BibTeX.
  157. ## License
  158. [MIT](LICENSE).
  159. > Outputs are drafts. Review by a domain expert is recommended before any citation,
  160. > submission, or decision. Verify numbers, citations, and claims.
  161. ## Acknowledgments
  162. Thanks to [linux.do](https://linux.do) — a vibrant tech community where this project is shared and discussed.

Ratings and comments

Sign in to rate this skill.

Sign in

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