tailor-to-venue
Skill59 starsChanged 5 months agoApache-2.0View on GitHub
Diffs a paper draft against a machine-readable venue profile and produces a four-part tailoring plan - contribution reframing for the chosen track (research vs applications/industry vs demo vs short/vision), a page-budget cutting plan, a template/documentclass switch plan (acmart, IEEEtran, NeurIPS-style, llncs), and an anonymization sweep matched to the venue's blind level. Use when the user wants to adapt, retarget, resubmit, or port a paper to a different conference, journal, or track; fit a paper into a page limit ("cut to 10 pages"); switch LaTeX templates ("convert IEEEtran to acmart"); anonymize for double-blind review; or asks "make this fit SIGSPATIAL/NeurIPS/ICDE/KDD", "which track should this go to", or "retarget my rejected paper".
In this repository43 other files it gives its agents.
- docs/llms-full.txtllms.txt
- docs/llms.txtllms.txt
- add-venue-profileSkill
- anonymize-paperSkill
- assess-paperSkill
- benchmark-paperSkill
- check-originalitySkill
- draft-related-workSkill
- draft-surveySkill
- fetch-paperSkill
- find-papersSkill
- fit-page-limitSkill
- literature-reviewSkill
- make-posterSkill
- make-slidesSkill
- match-styleSkill
- orchestrate-paperSkill
- paper-profileSkill
- parse-cfpSkill
- plan-submissionSkill
- polish-proseSkill
- polish-tables-figuresSkill
- preflight-checkSkill
- prepare-artifactsSkill
- prepare-camera-readySkill
- refactor-research-codeSkill
- refactor-structureSkill
- reflect-and-improveSkill
- reflect-paperSkill
- rehearse-qaSkill
- render-workspace-htmlSkill
- select-venueSkill
- simulate-reviewersSkill
- study-exemplarsSkill
- test-research-codeSkill
- triage-reviewsSkill
- verify-citationsSkill
- verify-claimsSkill
- verify-resultsSkill
- work-with-overleafSkill
- write-abstractSkill
- write-rebuttalSkill
- write-talk-scriptSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 4 are distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 62
Run from the repo root (or pass absolute paths):
- Distinctive: line 85
All three print Markdown and exit 2 with a clear message on bad input.
- Very specific: line 105
If the draft needs new citations from the target community, route every one through the verify-citations skill — never add an unverified …
- Very specific: line 110
Write tailoring-plan-<venue-id>-<track>.md next to the draft with sections:
- Distinctive: line 171
Never submit to any submission system on the user's behalf.
- Distinctive: line 172
Never state venue requirements from memory — profile + live-CFP verification only.
- Very specific: line 174
Never fabricate or hand-type citations.
- Very specific: line 176
Never plan template tampering (negative \vspace, margin/font tricks) to meet a page limit — these are desk-reject triggers.
- Very specific: line 178
Do not paste text from other authors' papers into the draft.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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
Tailor to Venue
Diff a draft against a target venue's requirements and produce a concrete tailoring plan: what to reposition, what to cut, what to reformat, and what to anonymize — before any edit is made. This skill plans; the user (or a follow-up request) executes the edits.
When to use
- Retargeting a paper (new submission, rejection, or venue switch) to a different conference, journal, track, or page limit.
- Converting between template families (acmart / IEEEtran / NeurIPS-style / llncs) or between blind levels (single ↔ double ↔ triple).
- Deciding which track at one venue fits the work best.
Related skills: parse-cfp (build a missing venue profile), select-venue
(choose the venue first), preflight-check (final desk-reject lint before
submission), prepare-camera-ready (after acceptance).
Inputs
- The draft: main
.texfile (the scripts resolve\input/\include); optionally the compiled PDF and.bibfiles. - Target venue profile:
venues/conferences/<venue-id>.yml(schema:venues/schema.yml; family defaults merge automatically fromvenues/families/). - Target track name, if the user has chosen one.
Process
1. Resolve the venue profile
Find the profile in venues/conferences/. If none exists, do NOT invent
requirements — run the parse-cfp skill against the venue's CFP URL to
create one, or proceed with only facts quoted live from the CFP.
2. Re-verify against the live CFP (mandatory)
Profiles go stale and a wrong page limit causes a desk reject. Fetch the
profile's cfp_url and re-verify before relying on anything: page limits and
exclusions for the chosen track, deadlines and timezone, blind level,
template/documentclass invocation, and required sections. Note in the plan
what was verified and when; if the live CFP contradicts the profile, the CFP
wins — flag the profile for update. If the CFP cannot be fetched, mark every
profile-derived fact "UNVERIFIED — confirm on CFP" in the plan.
3. Pick the track with the user
List the profile's tracks with their page limits (the venue diff report includes this table). If the user has not chosen, recommend one based on the work's strongest claim — see references/contribution-reframing.md for what each track rewards — and confirm before planning.
4. Run the deterministic diff
Run from the repo root (or pass absolute paths):
python3 skills/tailor-to-venue/scripts/venue_diff.py <main.tex> --venue venues/conferences/<id>.yml --track <Track>This reports template/option gaps, required-section gaps, abstract length, author-block vs blind level, page-limit context, and the venue facts (submission system, deadlines, LLM policy) to carry into the plan.
Then size the page budget:
python3 skills/tailor-to-venue/scripts/page_budget.py <main.tex> --venue venues/conferences/<id>.yml --track <Track>And scan anonymization at the venue's blind level (add --pdf <paper.pdf>
and --bib <refs.bib> when available):
python3 skills/tailor-to-venue/scripts/anon_sweep.py <main.tex> --venue venues/conferences/<id>.ymlAll three print Markdown and exit 2 with a clear message on bad input. Treat script outputs as signals: the compiled PDF and the live CFP are ground truth.
5. Build the four-part plan
Turn the script reports into prose plans using the references:
- Contribution reframing — retitle, re-abstract, rewrite contribution bullets and evaluation emphasis for the chosen track: references/contribution-reframing.md
- Page-budget cutting plan — ordered ladder from structural exports to line-level compression, with estimated savings per step and the forbidden moves listed: references/page-budget-cutting.md
- Template switch plan — only if venue_diff reports a class/option gap; per-direction breakage list and bibliography mapping: references/template-switching.md
- Anonymization sweep plan — fix strategy per leak category, plus the
beyond-the-PDF checklist and a
camera-ready-restore.mdfor removed content: references/anonymization-sweep.md
If the draft needs new citations from the target community, route every one
through the verify-citations skill — never add an unverified reference.
6. Deliver the plan
Write tailoring-plan-<venue-id>-<track>.md next to the draft with sections:
- Verification record — what was checked against the live CFP, when, and any profile contradictions found.
- Track fit and contribution reframing (with rewritten title/abstract/ contribution-bullet drafts marked as proposals).
- Page-budget cutting plan (current estimate → target, ordered cuts with estimated savings, restore list).
- Template switch plan (or "no switch needed").
- Anonymization sweep plan (findings with fixes, beyond-the-PDF checklist).
- Submission logistics — system, URL, deadlines with timezone, rebuttal format, LLM-policy disclosure needs.
Offer to execute the plan step by step; after edits, recommend
preflight-check as the final gate.
Worked example
A double-blind KDD draft was rejected; the user wants to retarget it to
SIGSPATIAL's Research track. Both venues use acmart sigconf, so this case
exercises three of the four plan parts (no template switch) and shows the
de-anonymization direction.
python3 skills/tailor-to-venue/scripts/venue_diff.py paper/main.tex \
--venue venues/conferences/sigspatial-2026.yml --track ResearchReading the reports against the live SIGSPATIAL CFP, the plan comes out:
- Verification. SIGSPATIAL Research is single-blind, 10 pages excl. references + appendix (KDD was double-blind, 8 pages) — both confirmed on the live CFP on the date of the diff.
- Track fit / reframing. Same Research track, so light reframing: re-aim vocabulary and baselines at the spatial-data community and add the community-standard baseline KDD reviewers did not expect (references/contribution-reframing.md, "Cross-venue repositioning").
- Page budget. The limit grew 8→10 pages, so this is a de-compression,
not a cut: restore the ablation moved to the appendix for KDD and the
examples trimmed for space.
page_budget.pyconfirms headroom. - Template switch. None —
venue_diffreports the class matches. - Anonymization (reverse sweep). The hard part. The draft is anonymized;
SIGSPATIAL requires author names. Run
anon_sweep.py paper/main.tex --venue venues/conferences/sigspatial-2026.yml— at single-blind it flags the leftover "Anonymous Author(s)" placeholder. Restore the real\author/\affiliation/\emailblock, drop theanonymous/reviewclass options, re-link the real repository, and restore first-person framing. Keep funding out until camera-ready unless the CFP asks for it (references/anonymization-sweep.md, "De-anonymizing").
The deliverable is tailoring-plan-sigspatial-2026-Research.md; no draft file
is touched until the user asks to execute a step.
Output
A reviewed, verification-stamped tailoring-plan-<venue-id>-<track>.md, plus
the three raw script reports on request. No draft files are modified by this
skill unless the user asks for execution of specific plan items.
Guardrails
- Never submit to any submission system on the user's behalf; stop at the plan.
- Never state venue requirements from memory — profile + live-CFP verification only; unverifiable facts are labeled UNVERIFIED.
- Never fabricate or hand-type citations; new references go through
verify-citations. - Never plan template tampering (negative
\vspace, margin/font tricks) to meet a page limit — these are desk-reject triggers. - Do not paste text from other authors' papers into the draft; exemplar
study happens transiently via the
study-exemplarsskill.
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.

