steven65026502
- 28 skills
- 0 followers
- 13 hours ago last updated
- ▌ DOCX Win · steven65026502 bundleUse native Microsoft Word COM on Windows for layout-sensitive DOCX work, tracked changes, comments, fields, tables of contents, headers, pagination, PDF export, and visual verification. Prefer this over cross-platform DOCX tooling when Word is installed and fidelity matters; use doc for simpler cross-platform work.
- ▌ Figma Use · steven65026502 bundleRequired safety and API rules for every available use_figma tool call. Use when the user requests programmatic Figma node, variable, component, layout, style, or file-context operations and use_figma is actually callable. If the tool or target file is unavailable, report the prerequisite instead of simulating a canvas change.
- ▌ Gh Fix CI · steven65026502 bundleUse when a user asks to diagnose or fix failing GitHub PR checks that run in GitHub Actions. Inspect checks and logs with gh and report evidence. If the user asked only for diagnosis, stop before edits; if they explicitly asked to fix, implement the focused repair without a redundant approval round. External providers remain out of scope.
- ▌ Notebooklm · steven65026502 bundleQuery and manage Google NotebookLM notebooks through the bundled browser automation when the user explicitly mentions NotebookLM or supplies a NotebookLM URL. Treat answers as source-grounded but still verify citations and report authentication, browser, or access failures; do not claim unsupported accuracy improvements.
- ▌ UI Styling · steven65026502 bundleImplement accessible application UI with shadcn/ui, Radix UI, and Tailwind CSS. Use for responsive layouts, forms, dialogs, tables, menus, component states, theme integration, and dark mode in a codebase using those tools. Do not use for posters, general art direction, or design-token architecture.
- ▌ Issue Ledger · steven65026502 bundleMaintain a layered issue ledger for coding, research, debugging, or project work. Use when the user asks Codex to record, organize, inspect, update, or summarize problems, solved items, unresolved issues, blockers, risks, evidence, decisions, verification results, or next steps across a project or repeated sessions.
- ▌ Research Hub · steven65026502 bundleOperate an installed research-hub CLI for source ingest, Zotero/Obsidian/NotebookLM workflows, cluster inspection, and vault maintenance. Use when the user explicitly asks to run a research-hub workflow or work in a repository configured for it. First verify the CLI and project configuration; otherwise explain the missing prerequisite instead of simulating success.
- ▌ Design System · steven65026502 bundleCreate or review design-token architecture and component specifications. Use for primitive-to-semantic-to-component tokens, CSS variables, spacing and typography scales, component states, and theme contracts. Do not use for general slide creation or one-off UI styling.
- ▌ Huashu Design · steven65026502 bundle用 HTML 製作高保真互動原型、動態視覺 Demo 或設計方向比較。適用於使用者明確要求 HTML 原型、可操作 mockup、動畫展示、設計變體或專家視覺評審；不適用於一般前端功能、PPTX 製作或只需文字建議的任務。影片、配音、音樂及浮水印只在明確要求時加入。
- ▌ UI UX Pro Max · steven65026502 bundleSearch the bundled UI/UX knowledge base for evidence-based style, color, typography, accessibility, chart, and stack-specific implementation guidance. Use when a web or mobile design decision or UI audit benefits from structured recommendations. Do not use for ordinary code changes that need no design judgment.
- ▌ Zotero Skills · steven65026502 bundleRead and, with explicit authorization, modify a Zotero library through the bundled dual local/Web API client. Use when the user explicitly asks to search Zotero, add or update items, notes, tags, or collections, or apply an approved cleanup plan. Reads may use the local desktop API; writes require Web API credentials and post-write verification.
- ▌ Humanizer Zh Tw · steven65026502 bundle去除繁體中文文字中的常見 AI 寫作痕跡，使語氣更自然且符合原作者目的。用於使用者明確要求自然化、去 AI 味、潤飾或審閱中文文字；不得改變事實、引用、專有名詞、承諾或作者立場。
- ▌ Paper Summarize · steven65026502 bundleFill per-paper Key Findings, Methodology, and Relevance sections in both Obsidian Markdown and Zotero notes after research-hub ingest. Use only for an existing configured cluster when the user requests per-paper summaries. Verify the CLI, model delegate, source abstract, and both write targets first; stop and report partial failures without claiming atomic success.
- ▌ Github Maintainer · steven65026502 bundleMaintain GitHub repositories with an industry-style workflow. Use when Codex is asked to organize a repository, clean GitHub structure, update README/CHANGELOG/docs, apply a project-specific maintainer profile, inspect git status before committing, avoid committing secrets or large generated files, write professional commit messages, push confirmed changes, or prepare repository handoff documentation.
- ▌ Paper Memory Builder · steven65026502 bundleConvert the user's manuscript draft, figures, and available Zotero metadata into .paper/claims.yml and .paper/figures.yml for later writing and audit work. Use when the user explicitly asks to build paper memory or extract claims. Do not edit the manuscript, invent missing evidence, or use this for cited-paper cluster summaries.
- ▌ Figma Generate Design · steven65026502 bundleCreate or update a full application screen in Figma from code or a description by reusing available design-system components, variables, and styles. Use only when the user wants a Figma canvas deliverable and both figma-use and the required Figma tools are callable; otherwise report the missing prerequisite.
- ▌ Research Hub Multi AI · steven65026502 bundleWrite .coord/multi_ai_plan.md for a user-requested workflow involving two or more available AI delegates. Use only when the user explicitly requests delegation or multi-agent coordination. Verify each delegate actually exists before naming it; do not require this router for a single-agent task and do not claim execution from a plan alone.
- ▌ Figma Implement Design · steven65026502 bundleTranslates Figma designs into production-ready application code with 1:1 visual fidelity. Use when implementing UI code from Figma files, when user mentions "implement design", "generate code", "implement component", provides Figma URLs, or asks to build components matching Figma specs. For Figma canvas writes via `use_figma`, use `figma-use`.
- ▌ Obsidian Github Backup · steven65026502 bundleBack up a local Obsidian vault to a private GitHub repository with safety checks. Use when the user asks to upload, sync, commit, push, or back up Obsidian notes or a vault to GitHub, especially when the backup must stay private and avoid committing API keys, private keys, local REST credentials, or other secrets.
- ▌ Playwright Interactive · steven65026502 bundleUse a persistent Playwright session through js_repl for rapid browser or Electron inspection, iterative UI debugging, and evidence capture. Use only when persistent interactive tooling is available and materially helps; otherwise use the playwright CLI skill or report the missing prerequisite.
- ▌ Research Design Helper · steven65026502 bundleRun a guided five-part study-design interview and write .research/design_brief.md. Use only when the user explicitly asks for a Socratic design session, a design brief, or to be walked through research framing. For a simple research-method question, answer directly without forcing the interview or writing files.
- ▌ Zotero Library Curator · steven65026502 bundleProduce a read-only Zotero cleanup audit for duplicate DOIs, missing tags, collection placement, and tag hygiene. Use only when the user explicitly asks for a Zotero audit or preview cleanup plan. Never perform CRUD; hand approved changes to zotero-skills and report when Zotero connectivity is unavailable.
- ▌ Academic Writing Skills · steven65026502 bundleEnd-to-end academic manuscript workflow for drafting, revision, reviewer response, figure-text consistency, claim-evidence audits, and pre-submission checks. Use this skill whenever a user asks for manuscript sections, paper revision, rebuttal letters, journal compliance, overclaim detection, GPT-style prose cleanup, figure captions, or evidence-backed academic writing. It is especially useful for multi-section papers where context, claims, figures, and reviewer comments must stay consistent across revisions.
- ▌ Standard Word Doc · steven65026502 bundleApply or repair this skill's standard Chinese Word document house style, including headings, lists, tables, fonts, spacing, and page numbers. Use only when the user asks for standardized Word formatting or this house format; use docx-win for general native Word work and docx-template-translator for a supplied template.
- ▌ DOCX Template Translator · steven65026502 bundleConvert LaTeX, PDF, or Markdown into a complete Word document that follows a user-supplied DOCX template. Use for thesis, dissertation, report, or institutional templates whose cover pages, declarations, numbering, captions, equations, citations, and layout require project-specific conversion and visual verification.
- ▌ Literature Triage Matrix · steven65026502 bundleTurn a user-provided paper list, Zotero collection, or Obsidian cluster into a comparison matrix at .research/literature_matrix.md. Use only when the user explicitly asks to compare papers by common dimensions or identify central papers; do not replace per-paper summaries or claim access to sources that are unavailable.
- ▌ Research Project Orienter · steven65026502 bundleRead existing .research manifest files and produce a read-only orientation memo covering the research question, datasets, current stage, entry points, evidence, and open questions. Use when the user asks to orient in a project that already has manifests. If they are absent, report that and suggest research-context-compressor; do not run it automatically.
- ▌ Research Context Compressor · steven65026502 bundleInspect a research repository and write a compact `.research/` workspace manifest (project_manifest.yml, experiment_matrix.yml, data_dictionary.yml) so future AI sessions can orient themselves without rescanning the whole repo. Use when the user asks to "compress this project context", "create a research manifest", or "save the project context for future agents".