Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/colt-workflow
Use when planning a COLT (Conference on Learning Theory) submission calendar backward from the single AoE paper deadline — proof-completion and verification milestones, writing and appendix-assembly passes, CMT logistics, the rebuttal window, decision aftermath, and PMLR camera-ready with conference presentation in early summer.
Passed 22 of 24 test cases with the skill, and 12 of 24 without it.
and 12 more in the full report
10 fixed12 passed either way2 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.
Our team is submitting a theoretical machine learning paper to COLT. Which online conference management system handles paper submission and paper records for COLT?
| Checked | Without skill | With skill |
|---|---|---|
| The response identifies CMT as the paper submission platform for COLT. | Missed | Passed |
The instructions your model reads when the skill loads.
Use this as the project-management layer for a COLT paper. Anchor everything to the live timetable: for the 39th edition the CFP (checked 2026-07-08) set one submission deadline — February 4, 2026, Anywhere on Earth, via CMT — for a conference running June 29 to July 3, 2026 in San Diego. Notification and camera-ready dates were not on the checked pages (待核实); take them from the CFP and acceptance email of your cycle.
colt-topic-selection); if the result is 80% proved, the clock question is
whether the remaining 20% is verification or invention — only the first is
schedulable.colt-reproducibility). Schedule it as real work with an owner per theorem — it
finds bugs at a rate authors never want to believe.colt-submission).colt-author-response).colt-camera-ready);
rejection → triage between fix-and-resubmit (next COLT / ALT) and re-routing.| Weeks before deadline | Milestone | Failure smell |
|---|---|---|
| 10+ | All main results proved at whiteboard level | "The lower bound should follow" — it does not, yet |
| 8 | Theorem statements frozen; ledger of claims opened | Statements still absorbing scope changes |
| 6 | Full proofs written in LaTeX, appendix-structured | Proofs alive only in notebooks and heads |
| 4 | Independent verification pass complete; bugs triaged | "We'll verify while writing" (you will not) |
| 3 | Body drafted: setup, results, proof overviews, known-vs-new table | Overview promises diverging from actual proofs |
| 2 | Internal mock review by an uninvolved theorist | No one external has read the model section |
test cases passed, vs 12 / 24 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.