1.3천건 · 관련도순
Framework de pré-treinamento visão-linguagem que conecta codificadores de imagem congelados e LLMs. Use quando você precisar de legendagem de imagens, resposta a perguntas visuais, recuperação imagem-texto ou chat multimodal com desempenho zero-shot de última geração.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Avalia LLMs em mais de 60 benchmarks acadêmicos (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use quando benchmarking de qualidade de modelo, comparação de modelos, relatório de resultados acadêmicos ou rastreamento de progresso de treinamento. Padrão da indústria usado por EleutherAI, HuggingFace e laboratórios principais. Suporta HuggingFace, vLLM, APIs.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Use this Canvas-owned clean-room skill for PDF work where layout, extraction, generation, forms, page operations, or visual verification matter. Trigger for reading PDFs beyond plain text, creating PDFs, merging, splitting, rotating, watermarking, extracting tables or images, filling forms, OCR handoff, and final PDF QA.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Use when the user asks to create, add, convert, publish, or deploy a blog post from a draft, notes, Markdown, outline, or topic. Turns a draft into a complete, SEO/GEO-optimized post wired into the site (template, blog index, sitemap, routing) with QA and deploy. Framework-agnostic — discover the project's own conventions first.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Use when work produces a deliverable, factual claim, dataset, code change, decision, publication, deployment, automation, or irreversible action whose failure would matter; when the user asks for a quality gate, acceptance criteria, QA, completeness, validation, audit, evidence, preflight, release readiness, or a definition of done; or before claiming completion on medium- or high-risk work. Automatically decide whether a formal gate is warranted, derive task-specific pass/fail criteria, gather evidence, and block unsupported completion. Make sure to use this skill even when the user does not say "quality gate" if consequential work needs acceptance criteria or completion evidence. Skip formal gating for trivial, reversible, low-impact requests.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Implement project-consistent UI through page classification, bounded references, semantic tokens, skeleton-first delivery, and responsive screenshot QA. Use for frontend redesigns, dashboards, forms, settings, mobile pages, screenshot recreation, visual polish, or generic-looking generated UI.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Choose the narrowest reliable browser surface across Codex Browser, signed-in Chrome, Playwright CLI, and platform scripts. Use for browser automation, logged-in workflows, UI inspection, screenshots, form filling, publish previews, or frontend QA.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Codex adapter for Garry Tan's full gstack methodology across CEO/product pressure, CTO architecture, design, engineering, QA, security, shipping, documentation, and retrospectives. Use when the 总控/CEO or 架构/CTO role needs gstack method routing, implementation-plan hardening, or the user says gstack should help a role.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Create, restyle, or rebuild Chinese academic PowerPoint presentations from papers, theses, reports, figures, notes, templates, screenshots, or visual mockups. Use for presentation planning, Template DNA extraction, slide mockups, template-based editable PPTX generation, editable reconstruction, rendering, and final QA.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Build a complete presentation deck from a brief. Generates storytelling plan, slide operations, renders the deck, and runs QA. Requires extracted template contracts from /slides-extract. Use when the user wants to create slides, build a deck, generate a presentation, write a strategy deck, or says things like "make me a 10-slide deck on X", "create a presentation about Y", "build slides for the board meeting".
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Convert literature reviews, paper folders, thesis drafts, journal-style reports, or research outlines into rigorous academic slide decks and review artifacts. Use when Codex must create, rebuild, or polish PPTX presentations from scholarly sources while preserving academic logic, citation-aware image/formula handling, editable layouts, figure explanations, template fidelity, and export-based visual QA.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Enforce traceability from natural-language intent to machine predicates, final observables, evidence, and owner-readable acceptance. Use when conducting any audit, review, QA pass, parity or compliance check, security assessment, code or architecture review, visual or behavioral comparison, data validation, system verification, completion gate, or claim that an implementation satisfies a human request.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Improve AI application with evaluation-driven development. Define eval criteria, instrument the application, build golden datasets, observe and evaluate application runs, analyze results, and produce a concrete action plan for improvements. ALWAYS USE THIS SKILL when the user asks to set up QA, add tests, add evals, evaluate, benchmark, fix wrong behaviors, improve quality, or do quality assurance for any Python project that calls an LLM model.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Bootstraps and runs a multi-agent development team with roles (dev, QA, DevOps), parallel work, and sprint plans that survive context resets. Use when kicking off a project with parallel AI agents, writing brainstorm prompts that produce real debate, planning sprints across chats, or recovering from context overflow mid-sprint.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Routes requests to the BMad Method (v6.12.4) agile framework: persona agents, planning skills (brief, PRFAQ, PRD, UX, spec, architecture, epics, sprint planning), shipping skills (build, code review, walkthrough, QA tests, retrospective), and core helpers (brainstorming, research, review, party mode). Use when the user mentions BMad or BMM, names a BMad agent such as Mary, John, Sally, Winston, or Amelia, or wants a structured flow from idea through PRD and architecture to built, reviewed code.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
AID Phase 4 - Development phase. Use for implementing features, TDD practices, code reviews, transitioning from planning to QA.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
AID methodology phase gate enforcement. Ensures work follows correct 6-phase order (Discovery -> PRD -> Tech Spec -> Impl Plan -> Dev -> QA & Ship). Enforces gates, validates transitions. Use before any work to check the current phase and refuse work belonging to a later phase.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Use for role-scoped FE/BE/DevOps/Admin/QA project docs under .codex/project-docs to preserve micro-context across long projects.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Use to wrap QA commands in ARB receipts for attestation evidence or to audit lint patterns from receipts.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Orchestrate multiple Antigravity skills through guided workflows for SaaS MVP delivery, security audits, AI agent builds, and browser QA.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Draft public Tamanu release notes for a given version. Use when the user wants release notes for a Tamanu release (e.g. "draft release notes for v2.45"). Pulls work from the specs and code that landed in that version's release/X.Y branch, plus the matching issues in the trackers (Linear's Tamanu team and Workhorse's Tamanu workspace), reconciles them, and writes docs/release-notes/vX-YY.md in the canonical v2.44 format for project managers and system administrators, plus a styled HTML rendering at .workhorse/design/designs/vX-YY-release-notes.html. Not for developer changelogs or QA test scoping (use scope-tamanu-release-tests for the latter).
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Produce a release test-impact analysis for a Tamanu version upgrade: compare two Tamanu releases and classify every functional area as Full Test, Smoke Test, or Skip, each with a short reason and the Linear card or PR reference. Use when the user wants a test plan, QA scope, regression scope, or "what to test" for a Tamanu release or upgrade, wants to compare or diff two Tamanu versions (e.g. v2.57 to v2.59), or asks which areas changed between two releases. Handles Tamanu's train-release model (release/X.Y branches vs vX.Y.Z tags) and maps changed files and PRs to Tamanu functional sections (Patient Management, Encounters/Charting, Surveys & Programs, Medications, Labs, Imaging, Appointments, Vaccinations, Sync, Settings, Reports, Admin/Permissions, Mobile, etc.). Optionally maps against a regression-testing spreadsheet's sections when one is provided.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Submission-grade remote-sensing Nature/high-impact journal figure workflow for Python or R. Use whenever the user asks to create, revise, audit, or polish manuscript figures, multi-panel scientific plots, map-led composites, remote-sensing image plates, sensor/product workflow figures, validation/error/uncertainty panels, or journal-ready SVG/PDF/TIFF outputs, especially for Nature-family, Remote Sensing of Environment, ISPRS, IEEE TGRS, or other high-impact venues. Before plotting, define the figure's conclusion, evidence logic, export needs, and review risks. If the user has not chosen Python or R, ask "Python or R?" and stop. Use only the selected backend for figure generation, previewing, exporting, and QA. Supports matplotlib/seaborn and ggplot2/patchwork/ComplexHeatmap. Not for dashboards or Illustrator/Figma-first infographics. Also trigger on general academic-writing figure needs even without the word "Nature", such as making figures/plots for a paper, scientific/academic plotting, data visualization for a manuscript, and Chinese phrasings like 遥感论文配图、学术写作配图、科研绘图、科研作图、地图出图、论文图表、可视化.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.
Generate a runnable exploit-validation command (Nuclei template, Metasploit module hint, AI-assisted Python script, or curl-based PoC) against a user-specified authorised target. Use when validating that a fix actually closed the vulnerability path, confirming a patch deployed correctly, or producing a copy-pasteable test command for QA. The skill never executes — the user runs.
한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.