ai-paper-writing
Skill0 starsChanged 29 days agoMITView on GitHub
Plan, draft, revise, polish, or compress AI, ML, NLP, and LLM research papers and individual sections. Use for manuscript prose and organization, including revisions based on reviews. Review-only requests belong to ai-paper-review; citation, build, and submission audits belong to ai-paper-check.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- HOWILLMAKEIT/skillstheir own version, first seen Nov 1, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- HOWILLMAKEIT/skills★ 9same 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.
- ai-paper-checkSkill
- ai-paper-reviewSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Read the shared principles, then only the guides needed below.
≈ “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.
Files
AI Paper Writing
Read the shared principles, then only the guides needed below.
Work from the supplied research
Identify the requested scope, manuscript files, intended reader, and available evidence. Reuse the existing outline, notation, bibliography, and author decisions. For a new paper or substantial reorganization, establish the paper plan before drafting. For a local edit, inspect the surrounding argument and relevant evidence without expanding into a full-paper process.
Introduce the question and intuition before dense notation. Each paragraph advances one step; explain each operation's purpose and connections. State novelty through a concrete technical difference. Replace defensive filler with scientific content, preserving real limitations. Interpret results rather than reciting table cells. Recommend missing experiments separately from completed results.
Load guidance by section or task
| Requested work | Reference and responsibility |
| Title or abstract | Title and abstract: communicate the task, contribution, and supported takeaway. |
| Introduction | Introduction: establish the gap and insight; select the contribution structure for the paper type. |
| Related work | Related work: position the contribution against the closest ideas and assumptions. |
| Setup or motivation | Setup and observations: define the task and justify the design from actual observations. |
| Method or supporting theory | Method: explain how the mechanism works and why each operation is needed. |
| Experimental results | Experiments: test the claims and interpret the evidence. |
| Conclusion or limitations | Conclusion and limitations: state the supported answer and its boundary. |
| Page allocation or compression | Page budget: prioritize experimental evidence for empirical papers. |
| Supplementary material | Appendix: organize by purpose, with proofs first when present. |
For a full draft, read the applicable section guides and allocation guidance. For one section, load its guide plus a neighboring guide only if a dependency or transition needs repair. Keep title, abstract, introduction, results, and conclusion consistent when authorized changes affect their shared claims.
Revise and deliver
Draft in the requested format; use LaTeX-ready prose when editing LaTeX. Retain valid labels, citation keys, equations, and values. Introduce new citations only after checking authoritative sources when available; otherwise mark the citation or claim as unverified. Do not turn a local prose task into a bibliography-wide audit.
After editing, check the affected argument for unsupported strengthening, term drift, missing reader context, unnecessary number repetition, and defensive filler. Keep real assumptions and counterevidence visible. For a requested review-and-rewrite, use ai-paper-review for diagnosis and implement the revisions here; do not repeatedly hand the task back and forth.
Use the existing project build command and engine after substantial LaTeX changes when available. Read the build guidance when changes affect compilation, references, or layout. Inspect the rendered PDF for layout work. Deliver the requested text or edits, briefly explain consequential changes, and identify unresolved evidence or verification. Report a draft as a draft when compilation or source verification remains unavailable.
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
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.