EYH0602

Revise academic paper text with colored markup tracking. Use when the user asks to "revise my paper", "address reviewer comments", "mark up changes", "do a revision pass", "prepare camera-ready with changes highlighted", or "self-revise the draft". Covers two revision modes: reviewer-requested revisions (wrapped in \rev for blue text) and self/agent-initiated revisions (wrapped in \selfrev for teal text). Use after the initial draft exists and changes need to be visible.

Systematic literature search and organization for academic papers. Use when starting a new paper's related work, expanding an existing Related Work section, surveying a research area, building a reading list, or when the user asks to "find papers on X", "what's the related work for X", "literature review", "survey this area", or "who else has worked on X". Also use when the user provides a draft and wants to strengthen its citations or find missing references to important prior work.

LaTeX formatting and compilation for academic papers. Use when the user asks about LaTeX templates, compilation errors, package conflicts, figure/table formatting, algorithm environments, math typesetting, or any "how do I do X in LaTeX" question in the context of a paper. Also use when setting up a new paper project, switching venue templates, debugging "undefined control sequence" or "missing $ inserted" errors, or when the user's paper has formatting issues visible in the PDF. Covers major ML/AI/NLP venues (NeurIPS, ICML, ICLR, ACL, EMNLP, CVPR, ICCV, ECCV, AAAI, IJCAI) and general LaTeX best practices.

Final-pass proofreading for academic papers before submission. Use when the user says "proofread my paper", "final check before submission", "check for typos", "is this ready to submit", "pre-submission checklist", or "last pass". Also use proactively as the final step in the paper-writing pipeline — after copy-editing is done and the paper is otherwise complete. Catches typos, numbering errors, cross-reference issues, notation inconsistencies, and formatting problems that other skills don't cover.

Verify citation integrity in academic papers before submission. Use when the user asks to "check my references", "verify citations", "clean up my bib file", "find missing citations", or "check for broken references". Also use proactively before any paper submission — when the user says "ready to submit", "final check", or "pre-submission review", reference checking should be part of that process. Handles BibTeX validation, \cite{} key matching, duplicate detection, completeness checks, and orphaned reference detection.

Academic English copy-editing for research papers. Use when the user asks to "copy-edit my paper", "improve the writing", "fix the English", "polish the language", "make it sound more academic", "tighten the prose", or "check grammar and style". Also use when the user is a non-native English speaker polishing a draft, when preparing a camera-ready version, or when reviewer feedback mentions "writing quality" or "language issues". This goes beyond grammar checkers by understanding academic writing conventions, hedging, and discipline-specific style.

Execute an approved implementation plan through subagent-driven development, committing each completed step for traceability, and opening a PR when done. Use when the user has a reviewed plan committed in the repo and says "implement", "build the plan", "execute the plan", "implement the plan", or wants to go from approved plan to pull request. Also use when the user references a plan file (e.g. `/implement path/to/plan.md`) and wants it implemented with quality gates. When a file path argument is provided, ALWAYS treat it as a plan to implement -- never as code to deploy directly.

Use when reviewing pull requests, before merging code, or when asked to review PR changes. Dispatches parallel specialized reviewers for bug detection, error handling, type design, test coverage, comment quality, and guidelines compliance.

How to install, link, and manage AI-agent skills with the `skillshub` CLI. Use this skill whenever the user wants to install a skill, add or remove a tap, share skills across multiple coding agents, sync `~/.skillshub/`, troubleshoot a missing or broken skill, write a `SKILL.md`, or do anything that touches the `skillshub` command — even when they describe the goal without naming the tool (e.g. "I want this skill in Cursor too", "why doesn't Codex see my skill", "package this folder as a skill").

Convert LaTeX math formulas from Draw.io files into clean, portable SVG files. Use this skill whenever the user wants to: export LaTeX/math formulas as SVGs, convert Draw.io formulas to SVG, get math equations out of Draw.io for use in other tools (Excalidraw, Lark, Google Slides, Canva, etc.), or mentions needing LaTeX formulas as images/vectors. Also use when the user mentions pdf2svg or drawio-to-svg conversion for math content.

Review Benchflow or SkillsBench task-run trajectories and integration-test Benchflow code changes. Use this skill whenever the user asks to audit traj health, failed or timed-out runs, healthy pass/fail/timeout status, no-skill leakage, skill loading, reward hacking, verifier isolation, metadata completeness, token usage, timing, Daytona-vs-Docker parity, path/root handling, coverage gaps, Docker/Daytona failures, or release-readiness of benchmark data.