Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

AI4S Agent (meta-skill)

A top-level orchestrator that turns a research direction or topic into a full package by chaining exploration, literature survey, experiments, and paper writing in order.

EducationAdvanced★ 170⑂ 17AI score 8/10Last updated: Jul 28, 2026

What it does

  • Orchestrates the pipeline: does no work itself; it invokes research-explorer → literature-survey → experiment-suite → paper-writer in the correct order.
  • Enforces a slug contract: derives one identical slug per topic (normalised text + 8-char SHA1) so every stage reads and writes output/<skill>/<slug>/latest/.
  • Reuses intermediate artifacts: seeds the paper's bibliography.bib from the survey's bib file and expands it to 200+ citations, pulls numbers from results.json, and copies publication-grade figures.
  • Keeps disclosure consistent: the simulated flag in results.json is the single source of truth for the simulated-data note in the report and the paper's \thanks, with a mandatory human-review clause in all cases.

Who it's for

  • Grad students and researchers in AI-for-Science who want a draft survey, runnable experiment code, and a paper in one pass.
  • Anyone developing or debugging this multi-stage research pipeline and comparing each stage's output.
  • Users who need automated bibliography freshness checks (≥60% from the last three years) plus LaTeX builds.

Examples

  1. Broad direction in: give "transformer time series forecasting"; the explorer produces a topic matrix, you pick one topic, and the remaining three stages run on it.
  2. Specific topic in: give "Transformer-based long-horizon forecasting with patch tokenisation"; exploration is skipped and survey/experiment/paper run directly.
  3. Real measured data: pass a path to your own results.json; the experiment stage uses it instead of simulating, and the paper drops the simulated-data clause.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/ai4s-research/ai4s-skills/HEAD/skills/ai4s-agent/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/ai4s-agent folder from the GitHub repo ai4s-research/ai4s-skills into my ~/.claude/skills/ai4s-agent/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/ai4s-research/ai4s-skills.git && cp -r ai4s-skills/skills/* ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and create the skills folder if needed: mkdir -p ~/.claude/skills
  2. Clone the repository: git clone https://github.com/ai4s-research/ai4s-skills.git
  3. Copy all skills over — the meta-skill cannot run alone: cp -r ai4s-skills/skills/* ~/.claude/skills/
  4. Verify with ls ~/.claude/skills that ai4s-agent, research-explorer, literature-survey, experiment-suite, and paper-writer are present.
  5. Install a TeX distribution (TeX Live or MacTeX) with latexmk so the survey and paper PDFs can build.
  6. Restart Claude Code and ask: "Use the AI4S agent to build a research package on <topic>."
  7. Collect deliverables from the output/ directory tree in your working folder.
View source on GitHub ↗License: MIT