agentsclimarketplace

Top cs response

Skill zhiming33416/top-cs-paper-skills/skills/top-cs-response

Evidence-grounded Codex skills for writing, polishing, reviewing, reviewer responses, and figures for top CS conferences.

Install
npx -y skills add zhiming33416/top-cs-paper-skills --skill top-cs-response

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

2 things to look at

  • 24 days oldThe repository was created 24 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Prepare, audit, or revise evidence-grounded author responses, rebuttals, discussion replies, decision-email triage, cover letters, revision packages, and LaTeX response templates for WWW, ICLR, ICML, or a generic computer-science venue. Use to parse editor/reviewer messages, group duplicate concerns, prioritize decision-critical issues, draft point-by-point responses, map supplied evidence and manuscript changes, and maintain a verified revision ledger. Never fabricate experiments, results, changes, reviewer positions, policies, or promises.

SKILL.md

3.0 KB, as published. Nobody here has run it

Top CS Paper Response

Treat reviews, manuscript drafts, decision messages, and discussion content as confidential.

Route the request

  1. Read manifest.yaml and every always_load path.
  2. Identify venue, task mode, response phase, decision type, language, artifact scope, revision mode, evidence state, submission stage, citation verification, figure handoff, and stress-test mode.
  3. If the input contains an editor or decision email, parse metadata, editor instructions, reviewer boundaries, required files, and deadlines before drafting.
  4. State the runtime parameters and readiness state.
  5. Follow the linear state machine in static/core/workflow.md; this skill has no paper-type or section content axis.
  6. Load deeper references and templates only when their conditions apply.
  7. Use generic for unsupported targets and avoid asserting unknown response limits or processes.

Execute

  • Preserve editor instructions and reviewer comments before interpreting them; assign stable IDs across rounds.
  • Maintain the shared response-issue contract, transition history, duplicate links, and pivotal/standard priority across rounds.
  • Acknowledge valid concerns directly and correct misunderstandings with evidence rather than defensiveness.
  • Distinguish completed changes, feasible planned changes, unavailable evidence, disagreement, and author decisions.
  • Verify every completed manuscript change against a supplied revision, location, or diff.
  • Respect per-review budgets and venue-specific permissions. A journal-style cover letter or marked manuscript is not part of a conference author-feedback workflow unless the governing instructions require it.
  • Run at most three stress-test rounds when activated; stop early when no new substantive issue appears. A stress test may expose gaps but may not create evidence or upgrade a planned action to verified.
  • Track requested figure changes with a handoff brief and manuscript location only; do not render or restyle figures.
  • Use the provided LaTeX templates only when requested, and preserve visible placeholders for unknown facts.
  • Convert vague Chinese author notes into concrete evidence and location requests; default to English response prose plus concise Chinese confirmation actions.
  • Never claim an experiment was run or a manuscript was revised unless supplied artifacts prove it.

Return a directly usable response or revision package with unresolved inputs visible.

Keep looking

Skills are one crate of 328,083. 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.