agentleFS
Sign inStart free
ScodiveScodive/Paper-Agent-Skills/skills/academic-paper-writing/SKILL.md

academic-paper-writing

Skill20 starsChanged 9 days agoMITView on GitHub

Revise academic manuscripts using supplied experiment records and bibliography, with anchored suggestions, exact edits and evidence checks. Also plan outlines and drafts with missing results explicitly marked.

  • Installs packages

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. Gavin-Tao/claude-skills★ 3same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository2 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 5 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use real manuscript text, experiment records and bibliography as the working materials.

    line 11
  2. Very specific:

    When this skill was installed with the skills CLI, its execution tools are bundled under scripts/.

    line 17
  3. Very specific:

    Use that Python executable and runner in place of python3 src/main.py below.

    line 27
  4. Very specific:

    Keep the skill installation separate from the paper's working directory.

    line 28
  5. Distinctive:

    Read the revision contract when preparing or responding to a revision package.

    line 35
  6. Very specific:

    Read the generated agent_task.md, complete manuscript.txt, all chunks and materials.json.

    line 42
  7. Distinctive:

    Do not invent experiments, numeric outcomes, citations or completed revisions.

    line 55
  8. Distinctive:

    Check whether evidence uses the same method, dataset, split, metric and units.

    line 58
  9. Distinctive:

    Treat writing patterns as contextual examples, never mandatory contribution counts, sentence lengths, or guarantees of publication quality.

    line 60

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Evidence-aware academic writing

Use real manuscript text, experiment records and bibliography as the working materials. Preserve supported technical content and separate hypotheses, planned experiments, user-reported results and independently verified evidence.

Standalone installation

When this skill was installed with the skills CLI, its execution tools are bundled under scripts/. Set SKILL_DIR to this installed skill directory. If needed, create a Python 3.10+ virtual environment and install its dependencies:

python3 -m venv .paper-agent-venv
.paper-agent-venv/bin/python -m pip install -r "$SKILL_DIR/scripts/requirements.txt"
.paper-agent-venv/bin/python "$SKILL_DIR/scripts/paper_agent.py" --help

Use that Python executable and runner in place of python3 src/main.py below. Keep the skill installation separate from the paper's working directory. The runner works outside the source repository. Do not reinstall if the active Python already provides the listed dependencies. An installed paper-agent CLI is also supported. No model API key is required by the execution tools.

Revision workflow

Read the revision contract when preparing or responding to a revision package. From a repository checkout:

python3 src/main.py prepare --paper paper.tex --experiments results.csv --references refs.bib --out runs/review-01

Read the generated agent_task.md, complete manuscript.txt, all chunks and materials.json. Write response.json matching response.schema.json, then:

python3 src/main.py apply --run runs/review-01 --response runs/review-01/response.json
python3 src/main.py check --run runs/review-01

With an installed distribution, replace python3 src/main.py with paper-agent. The scripts do not invoke a model; the host agent performs review and writing.

  • Anchor suggestions and claim assessments to original lines and evidence IDs.
  • Make focused edits; retain notation, methods and supported conclusions.
  • Do not invent experiments, numeric outcomes, citations or completed revisions.
  • Bibliographic metadata alone cannot establish source support. Read the source excerpt or mark the claim as needing evidence; a matching key is insufficient.
  • Check whether evidence uses the same method, dataset, split, metric and units. A matching number alone cannot establish support or statistical significance.
  • Treat writing patterns as contextual examples, never mandatory contribution counts, sentence lengths, or guarantees of publication quality.
  • Conference requirements must include venue, year, track, submission stage and official source. Do not rely on a timeless page-limit table.

Planning without completed evidence

outline --idea "..." --out runs/outline-01 and draft --outline outline.json --out runs/draft-01 export host-agent tasks. Leave results null or visibly marked [NEEDS_EVIDENCE]. These are planning artifacts, not evidence of completed work.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free