Use when drafting an ACM CCS rebuttal during the HotCRP author-response window, covering threat-model defenses, adaptive-attack objections, ethics and disclosure questions, anonymity constraints, adversarial reviewer pushback patterns, and decision-focused replies aimed at the meta-reviewer.
SkillScore73Maintained
Automatic
◆Research pack
Last verified 8/15/2026
What it changes
Passed 16 of 22 test cases with the skill, and 7 of 22 without it.
Without skill32%
With skill73%
Checks it passed that the model alone missed
The [Decision dimension] tag value is ethics.
The [Decision dimension] field is set to ethics.
The [Decision dimension] tag value is threat model.
and 9 more in the full report
9 fixed7 passed either way6 still failing0 broke
Pass-rate lift
+41 pp
Tokens
+16%
Agent turns
0%
Response time
-32%
Model gemini-3.6-flashMeasured 8/15/2026v116 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 9
Prompt
Reviewer 2 found a potential flaw in Lemma 3.2 of our formal cryptographic protocol proof. We found that adding one constraint on nonce freshness resolves the flaw without affecting the main theorem, as verified in Appendix B. Draft a HotCRP reply. Output five bracketed fields: Priority issue, Decision dimension, Draft response, Evidence anchor, and Forbidden content removed.
What was checked, without and with the skill
Checked
Without skill
With skill
The [Decision dimension] field is set to correctness.
Missed
Passed
The draft response proposes the nonce freshness constraint for Lemma 3.2.
Passed
Passed
case-10 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md62 linessingle file
CCS Author Response
Use this after CCS reviews are released. Reopen the current HotCRP rebuttal instructions and
any length or format limits before drafting, because response mechanics are set per cycle.
Triage
Answer concerns that affect the decision: threat-model soundness, novelty, correctness,
evidence quality, ethics, and disclosure. Skip cosmetic complaints unless space allows.
Use evidence already in the submission: sections, appendices, measurement tables, and the
ethics section. The rebuttal clarifies the submitted paper; it does not add a new paper.
Keep the reply anonymous. Do not reveal institution, authorship, private URLs, or the
identity of a disclosed vendor if that would deanonymize you.
Correct factual misreadings first, then address the strongest objection to the attacker
model or the adaptive evaluation.
Do not promise experiments you cannot run before camera-ready; promise only edits that add
no unsupported claim.
Drafting pattern
State the decision-critical correction or concession in the first line.
Point to exact submitted evidence (section, figure, appendix).
Explain the security consequence — why the attack still holds, or the defense still binds.
Offer a scoped camera-ready edit only where it clarifies without over-claiming.
Adversarial reviewer pushback patterns
Pushback
What it signals
CCS-ready fix
"The attacker assumes too much"
The threat model looks unrealistic
Point to where the capability is justified, or scope the claim to the realistic subset and concede the rest
"You never tested an adaptive attacker"
The defense evaluation looks incomplete
Reference the adaptive experiment if present; if absent, concede and commit to a bounded camera-ready addition only if honestly feasible
"This is already a known CVE"
A prior-art gap
Distinguish root cause, scope, or impact from the CVE, with the specific technical delta
"The ethics of this are unaddressed"
Disclosure or harm concern
State the disclosure timeline, the harm-minimization steps, and the reasoning, quoting the ethics section
Response micro-example
Reviewer objection: the key-recovery attack assumes co-located measurement that real
attackers cannot achieve. Reply skeleton:
Concede the strongest form needs co-location; scope the practical claim to that setting.