paper-outline
Skill1 starsChanged 3 months agoMITView on GitHub
Plan a scientific paper with the user by filling out the group's Outline document, one decision at a time.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- chaobo8484/MathX-math-model-skillstheir own version, first seen Oct 11, 2026
- kreid415/claude-science-skillstheir own version, first seen Oct 3, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- kreid415/claude-science-skills★ 1same name
- chaobo8484/MathX-math-model-skills★ 0same 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 repository39 other files it gives its agents.
- arch-mapSkill
- ask-meSkill
- autodevSkill
- codebase-designSkill
- code-reviewSkill
- context-barSkill
- contractSkill
- copilot-reviewSkill
- crapSkill
- deep-research-consolidateSkill
- deep-research-promptSkill
- domain-modelingSkill
- drop-reviewSkill
- fleet-reviewSkill
- framing-decisionsSkill
- grillingSkill
- grill-meSkill
- grill-with-docsSkill
- handoffSkill
- herdrSkill
- implementSkill
- mergedSkill
- meta-analysisSkill
- mutation-testSkill
- obsidian-vaultSkill
- probeSkill
- prSkill
- setup-matt-pocock-skillsSkill
- show-meSkill
- slack-askSkill
- slack-notifySkill
- sparSkill
- sync-matt-pocock-skillsSkill
- tddSkill
- to-specSkill
- triageSkill
- writing-for-agentsSkill
- writing-git-commitsSkill
- writing-great-skillsSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
Keep a ledger, visible to the user and updated as you go: one row per objective, carrying the subsection and key figure that earn it and …
- Distinctive: line 34
Read what the user hands you at invocation — that input is the brief, and it outranks anything you infer.
- Very specific: line 38
When Outline.md or Outline.docx already exists there, load it: this run sharpens that outline rather than opening a blank one.
- Very specific: line 48
Run the interview per grilling: one question at a time, AskUserQuestion with your recommendation first, and wait for the answer before the …
- Distinctive: line 65
Done when: every ledger row is closed, or parked with the user's explicit say-so.
- Very specific: line 69
Write Outline.md in the paper's folder, in the shape bin/render_outline.py documents, then render it:
≈ “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
Plan a new paper — or sharpen one already underway — by filling out the group's Outline document together with the user.
Two artifacts, side by side in the paper's folder: Outline.md, where the work happens, and
Outline.docx, what co-authors read.
The spine: every claim gets earned
A paper is a set of claims. The outline is where each one is earned — backed by a result the work actually produces — and closed — restated as a conclusion.
Keep a ledger, visible to the user and updated as you go: one row per objective, carrying the subsection and key figure that earn it and the conclusion that closes it.
| Objective | Earned by | Closed by |
| the claim, one sentence | subsection + key figure | conclusion |
An objective no result reaches is unearned — the work is missing, or the claim is too big for it. A result no objective reaches is orphaned — either it deserves an objective of its own, or it belongs in a different paper. Both are findings; put each to the user rather than quietly balancing the table yourself.
The ledger is what makes this session worth having. Reach it early and keep it on screen.
1. Ground yourself
Read what the user hands you at invocation — that input is the brief, and it outranks anything you infer. Then read the paper's folder: prior drafts, data, figures, the group's related papers.
When Outline.md or Outline.docx already exists there, load it: this run sharpens that
outline rather than opening a blank one.
Facts you can find, find. Decisions are the user's.
Done when you can state the scope, the objectives so far, and which results already exist — without having asked the user a single question of fact.
2. Grill
Run the interview per grilling: one question at a time, AskUserQuestion with your
recommendation first, and wait for the answer before the next one.
Walk the outline in the order below. Each row names what that section has to settle — arrive with a proposed answer drawn from your legwork, and grill the gap hardest of all, because an intro that misidentifies it costs the whole paper.
| Section | What to settle |
| Title | The claim a reader should carry away. |
| Scope | One sentence: what this paper covers, and what it deliberately leaves out. |
| Paper Objective(s) | The claims, as bullets. Each defensible with results in hand or planned. |
| Introduction | The funnel — broad, narrowing, state-of-the-art, the gap, then the goal. Name who came closest to the gap and why it is still open. |
| Methods | Each tool and method, and which objective's evidence it produces. |
| Results and discussion | One subsection per claim: the key figure, and what it shows that earns the claim. |
| Conclusions | One per objective, in the same order — the ledger's closing column. |
Done when every ledger row is closed, or parked with the user's explicit say-so.
3. Write and render
Write Outline.md in the paper's folder, in the shape bin/render_outline.py documents, then
render it:
python3 bin/render_outline.py <paper-folder>/Outline.mdThat fills the group template (assets/Outline-template.docx), so the headings, fonts, and
bullets match every other outline the group writes.
Done when both files sit in the paper's folder and Outline.docx carries the group's
headings.
4. Hand it back
Show the user the closed ledger and the two paths. Drafting the paper is a separate act — start it only when they ask.
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.