Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/cvpr-workflow
Use when planning a CVPR submission campaign end to end, covering the November abstract/paper/supplement deadline chain, coauthor reviewer-duty scheduling, the January rebuttal sprint, February decisions, camera-ready and June conference logistics, and the resubmission ladder across ICCV, ECCV, and the next CVPR.
Passed 21 of 22 test cases with the skill, and 10 of 22 without it.
and 19 more in the full report
12 fixed9 passed either way0 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 12 test cases the skill fixed. Same model, same prompt, with and without the skill.
We are building our project timeline for the CVPR 2027 cycle using published deadline dates from CVPR 2026. Audit our timeline assumptions and output the formal assessment using [Campaign stage], [Calendar], [Duty ledger], [Risk register], [Plan B], and [Cycle check].
| Checked | Without skill | With skill |
|---|---|---|
| Formats the cycle check section as current-year pages reopened on: <date>. | Missed | Passed |
| Notes that 2026 cycle dates serve as structural planning anchors until official 2027 dates release. | Missed | Passed |
The instructions your model reads when the skill loads.
CVPR is an annual machine with one intake: in the 2026 cycle, abstracts November 6, papers November 13, supplements November 20, rebuttal late January, decisions February 20, conference June 3–7 in Denver. This skill turns that spine into a project plan with owners, because CVPR deadline weeks are lost to coordination failures more often than to science failures.
| When | Milestone | Owner discipline |
|---|---|---|
| T–10 weeks (early Sep) | Freeze scope: claims, benchmarks, baselines | PI + first author sign one paragraph |
| T–8 weeks | Experiment matrix running; recipe ledger live | Every table row has a config |
| T–6 weeks | First full draft; flip-test the figures | Co-authors annotate, don't rewrite |
| T–4 weeks (early Oct) | OpenReview profiles audited for every author | One named owner; new coauthors close the list |
| T–1 week (Nov 6) | Abstract registered; profiles valid | Desk-reject exposure if missed |
| Nov 13 | Paper + CRF submitted | Freeze 24h early; upload before the panic hour |
| Nov 20 | Supplement uploaded | Videos/code packaged per cvpr-supplementary |
| Dec – Jan 12 | Coauthors deliver their assigned reviews | Late reviews endanger your paper |
| Jan 22–29 | Reviews land; one-page rebuttal | Pre-book this week; nobody travels |
| Jan 30 – Feb 5 | Discussion (no author access) | Resist mailing the AC |
| Feb 20 | Decisions | Both branches pre-planned |
| Spring (待核实) | Camera-ready + dataset release due | See cvpr-camera-ready |
| Jun 3–7 | Conference (workshops 3–4, main 5–7 in 2026) | Poster/talk/travel logistics |
Deadline fortnight (early November). The supplement trailing the paper by a week tempts teams to leave packaging "for week two" — then week two evaporates into video rendering and code scrubbing. Rule: the supplement plan (what goes in, who builds it) is written before the paper deadline; week two only executes it.
Rebuttal week (late January). Seven days, one page, all reviewers at once. Assign roles when reviews drop, not after a day of group grief:
# rebuttal-roles.yaml — filled within 12 hours of review release
reader: # summarizes all reviews neutrally, one page, no rebuttals yet
runner: # owns the 1-2 small requested experiments; hard stop at day 5
test cases passed, vs 10 / 22 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.