organize
Skill0 starsChanged 14 days agoMITView on GitHub
Structure a new scientific paper project before writing starts. Takes the target venue and the URL of its author guidelines, reads them to work out the venue's format (IEEE, ACM, Springer, Elsevier, ACS and others), fetches the official template, and scaffolds a LaTeX project that compiles from the first commit, with one file per section, shared macros, a build script and a project CLAUDE.md. Optionally publishes it to GitHub and connects Overleaf. Use when asked to set up, organize, scaffold or structure a paper project, or to start from a venue template.
16 other projects use a version of this.
Adopted by
Most recently taken up by
- willcoliveira/mac-cleanup-agenttheir own version, first seen Nov 15, 2026
- AndyMelnykov/obsidian-claudetheir own version, first seen Nov 15, 2026
- paulogabriel/beryltheir own version, first seen Nov 8, 2026
- SethGammon/Citadeltheir own version, first seen Nov 4, 2026
- zarubinvibe/athenatheir own version, first seen Nov 1, 2026
- snuffish/dotfilestheir own version, first seen Nov 1, 2026
Used with
Projects with it also use postmortem, dashboard and 2 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
Hongyu-yuSkillpostmortem
Use when analyzing failures, outages, incidents, or negative outcomes, conducting blameless postmortems, documenting root causes with 5 Whys or fishbone diagrams, identifying corrective actions with owners and timelines, learning from near-misses, establishing prevention strategies, or when user mentions postmortem, incident review, failure analysis, RCA, lessons learned, or after-action review.
Also in 3 projects that use this
GoogleSkilldashboard
Configures and launches the GKE Capacity & Triage Dashboard scoped to a GCP project ID, GKE cluster name, or target ML workload ID. Uses the configure-and-launch-dashboard action and launch-workload-dashboard workflow to verify local dashboard health, discover cluster capacity, trigger live scans, and generate deep-linked URLs. Use when the user types "/dashboard", says "launch dashboard", "open dashboard", "show capacity", or "view triage dashboard" with a project, cluster, or workload. Don't use for general workload log querying or automated failure attribution (use the troubleshoot skill).
Also in 3 projects that use this
ElasticSkillcreate-skill
Interactively create a new Claude Code skill and add it to the elastic-docs-skills catalog. Use when the user wants to generate a new skill, scaffold a slash command, or build automation for docs tasks.
Also in 3 projects that use this
genomewalkerSkilllearn
Learn from web sources, generate SKILL.md, and store key patterns in soul memory. Trigger keywords: learn, skill-gen, web-learn, documentation, tutorial.
Also in 3 projects that use this
Other versions16 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 16
- SethGammon/Citadel★ 927same name
- yunshu0909/yunshu_skillshub★ 760same name
- mcpware/cross-code-organizer★ 385same name
- mcpware/claude-code-organizer★ 383same name
- TyrealQ/q-skills★ 108same name
- sanghun0724/cmux-claude-skills★ 33same name
- seongjaeryu/bindly-claude-code★ 8same name
- mayankjain1815-commits/CITADEL★ 4same name
- AndyMelnykov/obsidian-claude★ 2same name
- zarubinvibe/athena★ 2same name
6 more not listed
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.
The file
We split this skill into the rules it gives an agent
9 rules. 0 are common, 6 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 13 edited versions we read keep these.
6
Distinctive
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
- Very specific: line 23
Check what already exists: If the working directory already contains a .tex file or a git repo, stop and ask before doing anything.
- Distinctive: line 70
Use the template link found in step 2, or ask the researcher for a URL or a downloaded zip.
- Distinctive: line 73
Treat a downloaded template as untrusted: download it into its own empty directory, inspect it, and copy only the class, style and example …
- Distinctive: line 101
Compile before you finish: If a TeX distribution is available, build once and report the result and page count.
- Very specific: line 104
Create paper-out/ and write STATE.md recording the venue profile, template provenance, and position as step 0, figures not yet selected, …
- Distinctive: line 110
Ask whether to publish. If yes, follow references/git-overleaf.md, which covers the two directions Overleaf supports, what its sync does …
- Distinctive: line 134
Never scaffold over an existing project: Stop and ask.
- Distinctive: line 135
Never impose a section structure: Ask; it depends on the venue and the kind of study.
- Very specific: line 144
If gh is unavailable, hand the commands over rather than improvising with credentials.
“Common” is measured: kept by at least 6 in 10 of the 13 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/organize
Step zero, before figures and before any writing. Produces a LaTeX project that follows the target
venue's format and that /write and /review can then work inside.
The local scaffold is the core job. Publishing it to GitHub and connecting Overleaf is an optional last step, and it creates a repository and pushes to it: those are outward-facing, so confirm name and visibility first and never create a public repository holding an unpublished manuscript unasked.
Before anything else
- Load the style rules. Invoke
paper-writing-style, and readreferences/session-state.mdsince this skill creates the initialpaper-out/. - Lock the language. Language of the user's first message, whole run.
- Check what already exists. If the working directory already contains a
.texfile or a git repo, stop and ask before doing anything. Scaffolding over an existing project is destructive.
Workflow
1. Get the venue and its guidelines
Take the venue from the argument, or ask. Then ask for the URL of its author guidelines or submission page, and what kind of paper this is (empirical study, experimental, review, short paper), since both decide the structure.
2. Read the guidelines into a venue profile
Fetch the page, and follow the links to the template and to any formatting rules. If you cannot fetch it (no network, a login wall, a PDF you cannot read), ask the researcher to paste the relevant text or attach the file. Extract only what the page says:
| Field | Why it matters |
| Document class and template, with version or year | Venues change requirements between editions |
| Page limit, and what counts toward it | An over-length paper is desk rejected unread |
| Abstract word limit and whether it is structured | Governs /write abstract |
| Review mode: single, double blind, open | Decides the \ifblind switch and the end matter |
| Citation and bibliography style | .bst file and \citep versus numeric |
| Required sections or order | Decides the src/ files |
| Mandatory declarations | Data availability, competing interests, CRediT, ethics |
| Figure rules | Format, resolution, colour, caption placement |
Present the profile as a short table with the source quoted or linked for each row, and mark every field the page did not answer as unknown. Ask the researcher to confirm or correct it. Do not fill an unknown from memory of what the venue family usually does: a convention you remember may be a year out of date, and a profile that sounds authoritative while being wrong is worse than a gap.
Common template families, as pointers for where to look, never as a substitute for reading the page:
| Venue family | Class |
| IEEE | IEEEtran |
| ACM | acmart |
| Springer LNCS | llncs |
| Springer journals | sn-jnl |
| Elsevier | elsarticle |
| ACS | achemso |
| APS | revtex4-2 |
3. Get the official template
Use the template link found in step 2, or ask the researcher for a URL or a downloaded zip. Never write a class file or approximate a template from memory. If neither exists, stop and say so.
Treat a downloaded template as untrusted: download it into its own empty directory, inspect it, and copy only the class, style and example files into the project. Never run anything it ships.
Record the venue, the template version and where it came from, so a later session can tell whether the template is current.
4. Confirm the plan
Present and get explicit agreement on:
- The venue profile from step 2.
- The
src/section files, derived from the profile and the kind of paper. Ask rather than impose: if the profile does not settle the structure, propose what the venue's accepted papers use and let the researcher change it. - The project directory name and local path.
- The
exemplars/folder, which is always created, with a.gitkeep. The researcher puts previous papers from the target venue and track there, and/draftbases the paper's structure on them. Say what it is for and that its contents are gitignored, so nobody commits other people's papers. - Whether to also publish to GitHub and connect Overleaf, which is opt-in.
5. Scaffold locally
Build the project as described in references/scaffold.md: layout, section files,
src/macros.tex, figures, exemplars/ with its .gitkeep, .gitignore, build script, and the project
CLAUDE.md. Feed the venue profile
into it: the blind switch, the page limit, the citation style and the mandatory declarations land in
main.tex, macros.tex and CLAUDE.md.
Compile before you finish. If a TeX distribution is available, build once and report the result and page count. If not, say the build is untested. A scaffold that does not build is worse than none.
Create paper-out/ and write STATE.md recording the venue profile, template provenance, and position as
step 0, figures not yet selected, plus a decisions.md entry for the section structure chosen. Name the
files created, and say once that paper-out/ is meant to be committed with the manuscript.
6. Offer Git and Overleaf
Ask whether to publish. If yes, follow references/git-overleaf.md, which covers the two directions Overleaf supports, what its sync does and does not do, and the exact steps. If no, say that a local repository alone is a perfectly good way to write a paper.
7. Hand off
Say what comes next and why:
The project is set up. The next step is not writing, it is selecting and ordering your figures. The figures tell the story and the text supports them, so results and discussion is written to a figure sequence that is already settled.
Once the figures are chosen, run
/ask-paperin this folder and it will take you through the rest.One more thing worth doing now: put previous papers from your target venue in
exemplars/. Download accepted papers from the same conference or journal and track, ideally 5 to 10 of the kind of paper you are writing. They are among the most useful things you can have when writing a new paper:/draftuses them to base your structure on what that venue actually accepts. The folder is never committed.
Hard rules
- Never write or approximate a venue class file. Get the official template or stop.
- Never fill a venue requirement from memory. Unknown stays unknown until the researcher or the page answers it.
- Never scaffold over an existing project. Stop and ask.
- Never impose a section structure. Ask; it depends on the venue and the kind of study.
- Never skip
src/macros.texor the projectCLAUDE.md. They make every later session start correctly. - Never create a GitHub repository without confirming name and visibility. Default private, and never public for an unpublished manuscript unless explicitly told.
- Never claim Overleaf sync works on a free account. It is premium only.
- Never tell the researcher to link an existing Overleaf project to an existing repository. It is not supported. Establish which direction applies first.
- Never commit raw data.
- Never emit an em dash (
---). - If
ghis unavailable, hand the commands over rather than improvising with credentials.
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.

