agentsclimarketplace

Ccs topic selection

Skill brycewang-stanford/Awesome-Journal-Skills/ACM-CCS-Skills/skills/ccs-topic-selection

Use when deciding whether a security project fits ACM CCS versus IEEE S&P, USENIX Security, NDSS, PETS, or a crypto/theory venue, identifying the security contribution type, and sharpening the threat model and attacker capability before writing begins.From its SKILL.md

Install
npx -y skills add brycewang-stanford/Awesome-Journal-Skills --skill ccs-topic-selection

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

3.7 KB, 757 tokens by cl100k_base, as published. Nobody here has run it

CCS Topic Selection

Use this before writing. ACM CCS is the SIGSAC flagship: it rewards work with a concrete attacker, a defensible threat model, and evidence that survives an adversarial program committee. Decide venue by community and contribution type, never by prestige ranking.

Fit test

  • Prefer CCS when the contribution is a broad computer-security result — a new attack class, a defense with measured cost, an applied-cryptography protocol, a systems or web-security mechanism, or a measurement study — aimed at the cross-area SIGSAC community.
  • Route to IEEE S&P (Oakland) when the work suits that PC's taste for foundational or systematization framing and the November cycle fits your calendar better.
  • Route to USENIX Security when the contribution is artifact-heavy systems security whose evidence lives in a runnable tool and open benchmark.
  • Route to NDSS when the core is network- and distributed-system security (protocols, DNS, routing, malware infrastructure).
  • Route to PETS/PoPETs when privacy is the primary lens rather than one property among many.
  • Route to CRYPTO/EUROCRYPT when the contribution is cryptographic theory whose proof, not its deployment, is the result.

Fit signal table

Signal in the projectCCS reading
New attack with a clearly bounded adversary and demonstrated impactCore fit — the house genre
Defense evaluated against adaptive attacks with deployment costCore fit
Applied crypto protocol with implementation and measured overheadCore fit
Internet-scale or ecosystem measurement with validated samplingCore fit
Pure cryptographic hardness proof, no systemCRYPTO/EUROCRYPT or a theory venue
Privacy-first metrics with no other security propertyPETS/PoPETs

Vignette: where a side-channel result goes

A project extracts keys from a deployed TLS library via a microarchitectural side channel, with a proof-of-concept exploit and a constant-time patch. CCS reading: strong fit — a concrete attacker, measured leakage, and a defense with overhead numbers is exactly the CCS arc. Strip the exploit and keep only an abstract leakage bound, and it drifts toward a crypto theory venue; expand the network-measurement of vulnerable hosts into the whole story, and NDSS becomes plausible; foreground only the privacy harm to users, and PETS fits better.

Sharpening moves before committing

  • Name the attacker: capabilities, knowledge, position, and what success means. If you cannot write the threat model in three sentences, the contribution is not yet CCS-shaped.
  • Decide the contribution type — attack, defense, protocol, measurement, tool, or study — because reviewers grade each against a different evidence bar.
  • Confirm the result fits the 12-page ACM sigconf body; CCS bodies are dense, and a paper needing 30 pages of proofs may belong at a journal or a theory venue.
  • Scope drifts across cycles; scan the current CFP topic list and recent proceedings before final routing.

Output format

[Fit] strong CCS / possible CCS / better elsewhere
[Best venue] CCS / IEEE S&P / USENIX Security / NDSS / PETS / crypto venue / other
[Contribution type] attack / defense / protocol / measurement / tool / study
[Threat model in one line] <adversary capability and goal>
[Top rejection risk] <threat-model / novelty / evidence / ethics / scope>
[Next action] <sharpen threat model, add evidence, reframe, or switch venue>

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.