Loading...
Loading...
Found 74 Skills
Builds a ranked venue shortlist for a research paper. Use when the user asks where to submit, which conference or journal to target, venue selection, track selection, or mentions CORE rank, h5-index, acceptance rate, CCF class, or submission deadlines. Maps the paper's topic and contribution type (method, system, dataset, demo, vision, industrial, survey) to venue-track pairs scored on topic fit, track fit, prestige signals, and deadline proximity computed from the machine-readable venues/ profiles — with mandatory re-verification of deadlines and page limits against each live CFP before the user relies on them.
Check whether reported paper metrics are consistent with outputs from the local code artifact. Use for result reproduction checks, artifact evaluation, reproducibility checklists, code-versus-paper audits, table or claim metric comparisons, and badge-readiness review.
Searches scholarly literature with zero API keys across DBLP, Crossref, Semantic Scholar, and arXiv. Use when the user wants to find papers, search the literature, list a conference's proceedings (e.g. "all SIGSPATIAL 2025 papers"), survey related work on a topic, look up papers by author/venue/year, get DOIs or citation counts, or check recent arXiv preprints. Ships a venue-alias table that solves cross-API venue naming (DBLP "conf/gis" vs Semantic Scholar "SIGSPATIAL/GIS" vs Crossref proceedings titles) plus polite, cached, 429-resilient search scripts. Trigger words - find papers, literature search, related work, proceedings, DBLP, Crossref, Semantic Scholar, arXiv, citation count, DOI lookup, venue search.
Deterministic desk-reject preflight linter for LaTeX paper submissions. Use it when a researcher asks "is my paper ready to submit", or mentions preflight, desk reject, submission check, page limit, anonymization, double-blind, missing checklist, or format/template compliance for a conference (NeurIPS, ICML, CHI, SIGSPATIAL, SIGMOD, ICDE, CVPR, LNCS...). Lints the .tex source against a machine-readable venue profile for documentclass and style-file options, margin/template tampering (geometry, savetrees, spacing hacks), anonymization leaks (author block, acknowledgments, repo links, first-person self-citations, \thanks, pdfauthor), required sections (NeurIPS checklist, ICML impact statement, AI-use acknowledgement, COI, CCS concepts/keywords), abstract length, keywords format, and page-limit risk. Every finding is reported with file:line. Runs bundled stdlib-only Python scripts; advisory only, never submits anything.
Drafts peer-review rebuttals and author responses in the three real venue formats - (a) strict one-page LaTeX rebuttal PDFs (CVPR-style), (b) OpenReview threaded per-review replies under 10,000-character budgets (NeurIPS/ICML/ICLR), and (c) revise-and-resubmit packages with a change-log response document (CHI, SIGMOD/ICDE revision rounds, journals). Anchors every response to paper sections, tables, figures, and line numbers; enforces character/page budgets and point-by-point coverage with bundled checker scripts; ships a one-page rebuttal LaTeX template. Use when the user says "write my rebuttal", "respond to reviewers", "author response", "reviews came back", "reviewer 2 says", "revise and resubmit", "response letter", "rebuttal deadline", or needs to fit a response into an OpenReview character limit or a one-page PDF. Drafts files on disk only; never posts, uploads, or submits.
Venue-calibrated peer-review simulation. Use when a researcher says "simulate reviewers", "mock review", "review my paper like a NeurIPS reviewer", "what would Reviewer 2 say", "red-team my paper", "find weaknesses before I submit", "peer review my paper", "strengths and weaknesses / pros and cons of my paper", "what to focus on", or wants a rubric score / borderline-reject risk estimate for a conference or journal (NeurIPS, ICML, ICLR, CVPR, KDD, SIGMOD, CHI, VLDB, LNCS, TKDE, TODS...). Builds a reviewer panel calibrated to the venue family and track (harsh NeurIPS main-track skeptics vs lenient demo-track judges), scores novelty/soundness/reproducibility/clarity on the venue's scale, hunts weaknesses grounded in quoted paper text, aggregates scores into a decision-risk band with borderline-reject flags, and outputs a prioritized fix list. Advisory only; improves the paper, never predicts the outcome, re-verifies facts against the live CFP, fabricates no citations or reviewers, and submits nothing.
De-AI-ifies and tightens academic prose without changing any technical claim. Use when a researcher says "this sounds like ChatGPT / AI-written", "humanize my paper", "remove the AI words", "too many delves and leverages", "polish my writing", "tighten the prose", "fix my hedging", "too many Moreover/Furthermore", "make my contributions active", or "keep my terminology consistent". Runs deterministic stdlib linters — prose_lint.py (leftover chatbot artifacts, LLM-tell vocabulary, connective stacking, em-dash density, double hedges, overclaiming boosters, passive contribution statements) and terminology_check.py (dataset/data set/data-set variants, acronym discipline, -ise/-ize mixing, glossary enforcement) — then drives a section-by-section edit pass using rewrite catalogs and the venue family's register norms. Numbers, results, claims, and citations are never altered; the venue's AI-use disclosure policy is surfaced and respected, never dodged.
Creates and checks conference research posters and short poster-session pitches. Use when the user asks to make a poster, prepare a research poster, adapt a paper into a poster, fit a poster to A0/A1/36x48in, check a beamerposter or tikzposter file, reuse paper figures, or write a 2-minute or 5-minute poster pitch. Bundled scripts resolve poster dimensions, inventory reusable figures, lint print-day problems, and time spoken pitches.
Writes a timed, word-budgeted speaker script for an exact conference-talk slot (e.g. a 12-minute talk plus 3-minute Q&A, or a 5-minute lightning talk). Use when a researcher mentions talk script, presentation script, speaker notes, conference talk, lightning talk, rehearsal, "what do I say on each slide", or "my talk is running long". Turns a slide deck or outline into spoken-register prose with a per-slide word budget at ~130 words/minute (calibratable), cumulative timing marks, written-out transitions, a marked cut list worth 10-15% of the talk for running long, and a Q&A handoff. Bundled stdlib-only Python scripts compute the budget (word_budget.py) and re-time any draft (script_timer.py with per-slide timing, cut savings, and an over/under verdict). Venue-aware but treats presenter instructions and the acceptance email as ground truth for slot length; never invents results — every claim spoken must come from the paper.
Verifies every citation in a BibTeX .bib file against Crossref, DBLP, Semantic Scholar, and arXiv to catch hallucinated, fabricated, or incorrect references before reviewers do. Use when the user says verify citations, check references, validate bibliography, check my .bib, fake or hallucinated citations, retracted paper, or DOI check — and before any draft, rebuttal, or camera-ready leaves the machine. Flags unresolvable entries (likely fabrications), author/year/venue mismatches, duplicates, and retractions, and points to the canonical record for each fix. Every skill that writes or edits citations must route new references through this gate.
Goal-driven conductor for the whole idea-to-published-paper lifecycle. Takes a research idea plus experiments and a target ("submit to NeurIPS 2026", or "help me pick a venue") and runs a goal->plan->execute->verify->reflect loop. It plans the stages, invokes the right lifecycle skills in order (literature-review, write-abstract, preflight-check, simulate-reviewers, and the rest), checkpoints with the author before each stage, and re-verifies the live CFP via parse-cfp rather than trusting a cached profile. Use when a researcher says "take my idea to a submitted paper", "run the whole pipeline", "orchestrate the paper for venue X", "what's the plan from here", or "what's done and what's next". Copilot, not autonomous author. It stops to ask at every stage gate, never fabricates results or citations, and never submits. Trigger words - orchestrate, pipeline, end to end, whole paper, idea to submission.
Prepares a reproducibility artifact (code/data) for submission and badging. Use when a researcher says "artifact evaluation", "artifact appendix", "reproducibility", "badge", "Artifacts Available/Evaluated/Functional/Reusable", "Results Reproduced/Replicated", "ACM badging", "USENIX/OSDI/SOSP AE", "SIGMOD ARI", "NeurIPS code/checklist", "ACL repro checklist", "Zenodo DOI", "Software Heritage", "anonymize my code/repo", or "package my code". Builds the artifact README + appendix, the dependency/run instructions, an anonymized repo for double-blind, and archival-DOI (Zenodo version vs concept) / Software Heritage SWHID guidance; resolves the ACM badge taxonomy and the Reproduced/Replicated era swap; and lints the artifact directory against the ML Code Completeness checklist with bundled stdlib-only scripts. Outputs an artifact-readiness checklist + packaging plan. Re-verifies the venue current artifact rules live. Advisory only; never submits.