Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/cav-review-process
Use when reasoning about how a CAV (Computer Aided Verification) submission is evaluated, covering the two-stage reviewing process (two reviews then an early-reject filter, then two more reviews with a rebuttal), the partial double-anonymity by category, the accept/reject outcome, the optional non-conditional artifact evaluation, and how CAV differs from TACAS and FMCAD.
Passed 16 of 22 test cases with the skill, and 12 of 22 without it.
and 8 more in the full report
6 fixed10 passed either way4 still failing2 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
Our paper received an early reject after the first round of CAV reviews due to an unconvincing evaluation benchmark. We want to submit the exact same manuscript directly to TACAS next week. Evaluate our plan and format the response using standard process assessment tags.
| Checked | Without skill | With skill |
|---|---|---|
| The response identifies the process stage tag as stage-1. | Missed | Passed |
| The response identifies resubmitting the paper unchanged as a forbidden move. | Passed | Passed |
| The response recommends reframing or addressing the gap before submitting to a sibling venue. | Passed | Passed |
The instructions your model reads when the skill loads.
Model the pipeline before interpreting any single review. CAV's process is a two-stage filter: a paper must survive the first two reviews before it reaches a rebuttal and the second pair of reviews. The most consequential mental shift for authors arriving from a single-round rebuttal conference is that a paper can be rejected before the rebuttal ever happens — so the first read has to stand on its own.
| Signal | What it means | Author move |
|---|---|---|
| Early reject after stage 1 | Two reviewers saw a fatal gap (unsound claim, weak evaluation, thin delta) | No rebuttal exists; reframe or reroute (TACAS/FMCAD/VMCAI) — do not resubmit unchanged |
| Passed to stage 2 | The contribution is plausible; specific concerns remain | Use the rebuttal to fix factual misreadings and supply the missing number/proof detail |
| Accept | Contribution, soundness, and evidence hold | Camera-ready + optional artifact; do not reopen scope |
| Reject after stage 2 | A concern the rebuttal did not resolve | Address it substantively before any resubmission |
The strategic reading: write the submission so its soundness and headline benchmark result are legible in the first two reviews. A contribution that only convinces after the rebuttal may never reach the rebuttal.
test cases passed, vs 12 / 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.