“code generation”的搜索结果

Elite website image-to-code skill for Codex. For visually important web tasks, it must first generate the design image(s) itself, deeply analyze them, then implement the website to match them as closely as possible. In Codex, it must prefer large, readable, section-specific images instead of tiny compressed boards, generate fresh standalone images for sections or detail views instead of cropping old ones, avoid lazy under-generation, avoid cards-inside-cards-inside-cards UI, and keep the hero clean, spacious, readable, and visible on a small laptop.

中文简介正在准备中。当前显示原文说明。

Use this skill whenever the user asks to generate, edit, inpaint, restyle, or create bitmap images through OpenAI's Responses API image_generation tool from Codex. This skill calls the Responses API, saves generated base64 image results to files, uses no npm or pip dependencies, works on Linux/macOS/Windows with either Node.js 18+ or Python 3, and reuses Codex's configured API base URL, model, and key instead of asking the user for credentials.

中文简介正在准备中。当前显示原文说明。

Use when the task involves generating, modifying, debugging, validating, or reviewing code. Do not use for non-code tasks.

中文简介正在准备中。当前显示原文说明。

Elite website image-to-code skill for Codex. For visually important web tasks, it must first generate the design image(s) itself, deeply analyze them, then implement the website to match them as closely as possible. In Codex, it must prefer large, readable, section-specific images instead of tiny compressed boards, generate fresh standalone images for sections or detail views instead of cropping old ones, avoid lazy under-generation, avoid cards-inside-cards-inside-cards UI, and keep the hero clean, spacious, readable, and visible on a small laptop.

中文简介正在准备中。当前显示原文说明。

Loads org- and repo-level coding rules from Qodo before code tasks begin, ensuring all generation and modification follows team standards. Use before any code generation or modification task when rules are not already loaded. Invoke when user asks to write, edit, refactor, or review code, or when starting implementation planning.

中文简介正在准备中。当前显示原文说明。

Loads org- and repo-level coding rules from Qodo before code tasks begin, ensuring all generation and modification follows team standards. Use before any code generation or modification task when rules are not already loaded. Invoke when user asks to write, edit, refactor, or review code, or when starting implementation planning.

中文简介正在准备中。当前显示原文说明。

Elite website image-to-code skill for Codex. For visually important web tasks, it must first generate the design image(s) itself, deeply analyze them, then implement the website to match them as closely as possible. In Codex, it must prefer large, readable, section-specific images instead of tiny compressed boards, generate fresh standalone images for sections or detail views instead of cropping old ones, avoid lazy under-generation, avoid cards-inside-cards-inside-cards UI, and keep the hero clean, spacious, readable, and visible on a small laptop.

中文简介正在准备中。当前显示原文说明。

Marp形式プレゼンテーションの専門レビュー・評価・改善スキル。 10軸スコアリング（100点満点）で品質を定量評価し、具体的な改善コード付きで提案する。 Use when the user asks to: - Review, evaluate, or critique a Marp presentation / プレゼンをレビュー / 評価 / 添削 - Improve, polish, or upgrade existing slides / スライドを改善 / ブラッシュアップ - Check slide quality, score slides, or audit a deck / スライドの品質チェック / 採点 - "このプレゼンどう？" "スライドを見てほしい" "プレゼン資料のフィードバック" Reads .md files with Marp frontmatter, scores across 10 dimensions, and outputs a detailed review report with concrete before/after code improvements. Works as a companion to the marp-slides skill (generation → review → improvement cycle).

中文简介正在准备中。当前显示原文说明。

Create Kotlin Compose design tokens from one of three supported sources — Google Stitch (MCP), Figma (MCP or JSON), or Codex Design (kta-design-spec output, Codex HTML/markdown artifacts). Three-layer architecture (Primitive/Semantic/Component) with code generation. Use when user asks to "generate design tokens", "create theme from Figma/Stitch/Codex design", "build Compose tokens from design spec", or "convert design source to tokens". Refuses any other input source and asks user for correct seed data.

中文简介正在准备中。当前显示原文说明。

Expert en conception de compilateurs (lexing, parsing, AST, code generation, LLVM, interpreters)

中文简介正在准备中。当前显示原文说明。

Guarantee valid JSON/XML/code structure during generation, use Pydantic models for type-safe outputs, support local models (Transformers, vLLM), and maximize inference speed with Outlines - dottxt.ai's structured generation library

中文简介正在准备中。当前显示原文说明。

Elite website image-to-code skill for Codex. For visually important web tasks, it must first generate the design image(s) itself, deeply analyze them, then implement the website to match them as closely as possible. In Codex, it must prefer large, readable, section-specific images instead of tiny compressed boards, generate fresh standalone images for sections or detail views instead of cropping old ones, avoid lazy under-generation, avoid cards-inside-cards-inside-cards UI, and keep the hero clean, spacious, readable, and visible on a small laptop.

中文简介正在准备中。当前显示原文说明。

Elite website image-to-code skill for Codex. For visually important web tasks, it must first generate the design image(s) itself, deeply analyze them, then implement the website to match them as closely as possible. In Codex, it must prefer large, readable, section-specific images instead of tiny compressed boards, generate fresh standalone images for sections or detail views instead of cropping old ones, avoid lazy under-generation, avoid cards-inside-cards-inside-cards UI, and keep the hero clean, spacious, readable, and visible on a small laptop.

中文简介正在准备中。当前显示原文说明。

Elite website image-to-code skill for Codex. For visually important web tasks, it must first generate the design image(s) itself, deeply analyze them, then implement the website to match them as closely as possible. In Codex, it must prefer large, readable, section-specific images instead of tiny compressed boards, generate fresh standalone images for sections or detail views instead of cropping old ones, avoid lazy under-generation, avoid cards-inside-cards-inside-cards UI, and keep the hero clean, spacious, readable, and visible on a small laptop.

中文简介正在准备中。当前显示原文说明。

Loads org- and repo-level coding rules from Qodo before code tasks begin, ensuring all generation and modification follows team standards. Use before any code generation or modification task when rules are not already loaded. Invoke when user asks to write, edit, refactor, or review code, or when starting implementation planning.

中文简介正在准备中。当前显示原文说明。

将颜色、间距和字体整理为共用的设计变量，生成CSS变量，定义界面组件在不同状态下的规范，检查代码中直接写入的值，并制作遵循同一品牌规则的HTML演示文稿。

• 统一颜色、间距和字体
• 整理按钮在不同状态下的规范
• 检查代码中直接写入的样式值

Build and troubleshoot TorchDrug 0.2.1 workflows for molecular property prediction, molecule generation, protein modeling, and knowledge graph reasoning. Connect datasets, models, training tasks, and evaluation while checking version compatibility and common implementation errors.

中文简介正在准备中。当前显示英语简介。

• Training and evaluating molecular property prediction models
• Building molecule generation workflows
• Learning representations of protein sequences and structures

Generate, compose, and edit product images, posters, illustrations, and social graphics through the paid Beatra service. The workflow covers reviewing upload files and estimated costs before each paid call, then checking results and actual charges.

中文简介正在准备中。当前显示英语简介。

• Creating product images
• Designing posters and social graphics
• Editing reference images and making photo variations

Create and maintain OpenAPI 3.1 specifications from existing code or initial designs. Use these API descriptions to prepare documentation, generate client SDKs—libraries for calling an API—and check whether implementations match their specifications.

中文简介正在准备中。当前显示英语简介。

• Create API documentation from existing code
• Define an API specification before implementation
• Generate client SDKs from API specifications

Write or revise instructions, specifications, tickets, and skill files so an agent can act without prior chat history. Make rules checkable, explain their reasons, define completion criteria, and keep authoritative information in one place.

中文简介正在准备中。当前显示英语简介。

• Clarifying instructions in CLAUDE.md
• Making specifications and tickets actionable
• Adding completion criteria to skill procedures

Add a diffusion model or Diffusers generation pipeline to SGLang. Examine reference code, choose a suitable pipeline structure, reuse existing components, and implement model settings and processing stages, with guidance for later multi-GPU support.

中文简介正在准备中。当前显示英语简介。

• Adding a new diffusion model to SGLang
• Porting a Diffusers generation pipeline
• Adding a variant by reusing existing model components

Build and debug sglang tests comparing token probabilities during prompt processing and step-by-step generation. Separate numerical differences from reused-state bugs, choose test thresholds, and locate the first operation that diverges.

中文简介正在准备中。当前显示英语简介。

• Adding probability consistency tests for a model
• Choosing or justifying a kl_div threshold
• Separating numerical differences from cache state bugs

Guide reinforcement learning, training models through rewards, with slime. Prepare data, configure Megatron-LM training and SGLang response generation, and set up asynchronous runs or custom multi-turn workflows for supported language models.

中文简介正在准备中。当前显示英语简介。

• Preparing reward-based training for GLM or Qwen models
• Overlapping response generation and training
• Training agents that use tools across multiple turns

Review code for privacy risks, prepare data protection impact assessments, and track personal data requests and deadlines. The skill supports GDPR and German privacy compliance work, with final compliance decisions referred to a data protection officer or legal counsel.

中文简介正在准备中。当前显示英语简介。

• Checking a service's code for privacy risks
• Preparing a data protection impact assessment
• Tracking deadlines for access and erasure requests