cli

Audio generation skill. Automatically selects the best dlazy CLI audio/TTS model based on the prompt. 音频生成技能。根据提示词自动选择最佳的 dlazy CLI 音频/TTS 模型。

Generate/edit images with Nano Banana Pro. Supports text-to-image and image-to-image. 使用 Nano Banana Pro 生成/编辑图片，支持文生图与图生图。

Generate/edit high-quality images with Nano Banana 2.0. Supports text-to-image and image-to-image. 使用 Nano Banana 2.0 生成/编辑高质量图片，支持文生图与图生图。

Chat with the dlazy sandbox agent — a project-scoped assistant that runs skills end-to-end over multiple turns. Discover skills and projects with dlazy skills list / dlazy projects list. 与 dlazy 沙箱 agent 对话 —— 一个以项目为单位、可端到端运行技能的多轮助手。用 dlazy skills list / dlazy projects list 发现可用技能与项目。

Anthropic's flagship multimodal model. Strong at long-context reasoning, code generation, and complex tool orchestration. Supports text, image, and video inputs. Anthropic 旗舰多模态模型，擅长长上下文推理、代码生成与复杂工具编排。支持文本、图片与视频输入。

Synthesize text into natural and fluent speech using Doubao TTS. 使用豆包 (Doubao) TTS 文本转语音模型，将文字合成为自然流畅的语音播报。

Calls the Apify YouTube downloader actor for a single video URL and returns one video file URL (uploaded to object storage) plus metadata (title, duration, width, height). Typical flow: use get-trends-videos to discover a pageUrl, then pass it to this tool to grab the file. 调用 Apify YouTube 下载 Actor，根据视频页 URL 拉取视频文件，返回单条已上传到对象存储的视频 URL 及标题、时长、分辨率等元数据。通常配合 get-trends-videos 使用：先搜索热门视频拿到 pageUrl，再用该工具下载选中的视频。

ElevenLabs eleven_v3 multi-voice dialogue: assign a different voice per line (up to 10) and render the whole conversation in one shot. Supports audio tags like [giggling], [whispers] — great for character dialogue, podcasts, and short skits. Before picking a voice, you can search for the right one via elevenlabs-search. ElevenLabs eleven_v3 多人对白合成：为每行台词指定不同音色（最多 10 个），一次性生成完整对话音频。支持 [giggling]、[whispers] 等情绪标签，适合角色对白、播客与短剧。选择音色前，可以从 elevenlabs-search 检索合适的音色。

ElevenLabs music_v1 model — generates 10–300s original music from a natural-language prompt. Good for BGM, ads, and short-video soundtracks. ElevenLabs music_v1 音乐生成模型，根据自然语言提示生成 10–300 秒原创音乐。适合 BGM、广告配乐与短视频音轨。

Search the ElevenLabs voice library by keyword, source, and category. Returns a playable preview for each matched voice so you can pick the right one before running TTS. 搜索 ElevenLabs 人声库：按关键词、来源、分类筛选可用音色，返回每个音色的试听样本，便于挑选后用于 TTS 配音。

ElevenLabs text-to-sound model — generates 1–22s short sound effects from a description. Suitable for foley, ambience, alerts, and game SFX. ElevenLabs 文本生音效模型，根据描述生成 1–22 秒短音效。适合拟音、环境声、提示音与游戏音效。

ElevenLabs scribe_v1 speech-to-text with auto language detection and optional speaker diarization. Suitable for subtitles, transcription, and meeting notes. ElevenLabs scribe_v1 语音转文字，支持自动语种识别与说话人分离，适合字幕、转录与会议记录。

ElevenLabs eleven_v3 text-to-speech with 12 curated multilingual voices and stability/similarity/style controls. Great for dubbing, audiobooks, and character dialog. Before picking a voice, you can search for the right one via elevenlabs-search. ElevenLabs eleven_v3 文本转语音，提供 12 种精选英文/多语种音色，支持稳定性、相似度、风格控制。适合配音、有声内容与角色对话。选择音色前，可以从 elevenlabs-search 检索合适的音色。

ElevenLabs Instant Voice Cloning (IVC). Upload a clean voice sample to clone a custom voice usable with ElevenLabs TTS. ElevenLabs 即时音色克隆（IVC），上传一段干净人声样本即可复刻自定义音色，可用于 ElevenLabs TTS 配音。

Executes a plan's flat shape list end-to-end. Each shape is dispatched through /api/ai/tool in dependency order (resolved from `shape://name:<X>` refs); upstream outputs are substituted into downstream input, and `promptRefs` text is prepended to the consumer's `prompt`. IMPORTANT: only shapes whose `props.status` is `idle` actually run — any other status (`completed` / `failed` / `processing`) is treated as already-done and its existing `props.url` is reused as the producer output. To resume after a failure, set the failed shapes (and any you want re-run) back to `idle`, optionally edit their `props.input`, then call execute again. 执行 plan 输出的扁平节点列表。按 `shape://name:<X>` 引用计算依赖顺序，依次通过 /api/ai/tool 调度每个 shape；上游产物的 url 会替换到下游 input 中相应字段，`promptRefs` 引用的文本会被拼接到消费者的 `prompt` 前面。重要：只有 `props.status === "idle"` 的 shape 才会真正执行；其它状态（`completed` / `failed` / `processing`）一律视为已完成，其当前的 `props.url` 被作为产物供下游引用。若上次执行中途失败，可将失败节点（以及任何需要重跑的节点）的 status 改回 `idle`，可选地修改 `props.input`，再次调用 execute 即可从失败位置继续。

ppt to video, word to video, excel to video, pdf to video, document to video — parse, outline, storyboard, voiceover, build, validate. Use when the user gives a document and wants an explainer, report broadcast, courseware, or training video. PPT 转视频、Word 转视频、Excel 转视频、PDF 转视频、文档转视频——解析、大纲、分镜、配音、合成、校验。当用户给一份文档、想要讲解视频 / 报告解读 / 课件 / 培训视频时使用。

Alibaba Bailian Fun-ASR recording transcription. Supports Chinese, English and other languages, with auto language detection and speaker diarization. Suitable for subtitles, transcription, and meeting notes. 阿里云百炼 Fun-ASR 录音文件识别，支持中英文及多语种，自动语种识别与说话人分离。适合字幕、转录与会议记录。

Generate multilingual, highly natural audio using Gemini 2.5 text-to-speech. 使用 Gemini 2.5 强大的文本转语音能力，生成多语言、高自然度的音频。

Google's multimodal model with strong long-context and vision understanding. Suitable for document parsing, image/video understanding, and structured output. Google 多模态大模型，长上下文与视觉理解强，适合文档解析、图像/视频理解与结构化输出。

A comprehensive generation skill. Can generate images, videos, and audio by automatically selecting the appropriate dlazy CLI model. 综合生成技能。能够根据用户意图自动选择合适的 dlazy CLI 模型来生成图片、视频或音频。

Fetches trending videos from YouTube or TikTok via Apify and returns up to N items as metadata (title, author, views, thumbnail, page URL, etc.) filtered by keyword, time window and sort order. Returns metadata only — does not download files. To get a playable video file for a chosen result, pass its pageUrl to download-trends-videos. 调用 Apify 抓取 YouTube 或 TikTok 的热门视频列表，按关键词、时间窗口与排序返回 N 条视频的元数据（标题、作者、播放量、缩略图、页面链接等）。仅返回元数据，不下载文件；如需获取可用的视频文件，请把选中的 pageUrl 传给 download-trends-videos。

OpenAI's general-purpose model balancing reasoning quality with response speed. Suitable for chat, writing, planning, and multimodal analysis. OpenAI 通用大模型，平衡推理质量与响应速度，适合对话、写作、规划与多模态分析。

GPT Image 2 model for text-to-image and image editing. Supports generating images from text as well as editing and synthesizing images with reference inputs. GPT Image 2 图像生成与编辑模型。支持文生图，以及通过提供参考图进行图像编辑和合成。

Efficient text generation, dialogue QA, and logical reasoning using Grok 4.2 text model. 使用 Grok 4.2 文本大模型，进行高效的文本生成、对话问答与逻辑推理。

Happy Horse 1.0 video model — one model covers text-to-video (t2v), first-frame-to-video (i2v), reference-to-video (r2v), and video editing (edit). The selected mode is automatically routed to the matching sub-model. Happy Horse 1.0 视频模型，一站式覆盖文生视频（t2v）、首帧生视频（i2v）、参考图生视频（r2v）与视频编辑（edit）：根据所选模式自动路由到对应子模型。

HeyGen Lipsync Speed: Fast lip-sync model, ideal for scenarios requiring rapid generation. HeyGen Lipsync Speed：快速唇形同步模型，适合对生成速度要求较高的场景

Turn a user's idea into a full video pipeline: story, characters, 3-view portraits, scenes, shots, keyframes, shot videos, and final concat. Emits a plan template for confirmation, then executes. 把一句创意扩展成完整的成片流程：故事 → 角色 → 三视图人设 → 场景 → 分镜 → 关键帧 → 分镜视频 → 拼接成片。先产出 plan 模板供确认，再执行生成。适合创意一键生成视频、idea 转视频。

A professional product image generation skill purpose-built for the Amazon e-commerce platform. Outputs comply with Amazon's image guidelines while optimizing for click-through and conversio 专为亚马逊电商平台打造的专业产品图生成技能，产出符合亚马逊图片规范，同时优化点击率与转化。适合亚马逊商品主图、白底图、场景图与详情图。

Image generation skill. Automatically selects the best dlazy CLI image model based on the prompt. 图片生成技能。根据提示词自动选择最佳的 dlazy CLI 图片生成模型。

A complete workflow skill for marketing brochure design, covering everything from requirements gathering, layout design, to mock-up delivery. It uses a 'layout-first + mandatory confirmation 营销宣传册 / 单页设计完整工作流技能，覆盖需求收集、版式设计到样稿交付，采用版式优先 + 强制确认的流程。适合宣传册、折页与营销物料设计。

Image replicate tool: analyzes the visuals, composition, colors, lighting, and style of the source image, builds a replicate prompt, and hands it off to Seedream 4.5 to generate a new image in the same style. 图像仿写工具：分析原图的视觉效果、构图、色彩、光影和风格，生成重绘提示词，并交给 Seedream 4.5 生成具有相同风格的新图像。

A structured workflow skill dedicated to social-media carousel design. The core method is 'decide intent first, then execute,' using a 'single-confirmation + cover-first' two-phase flow. 社交媒体轮播图 / 多图设计工作流技能，核心方法是先定意图再执行，采用单次确认 + 封面优先的两阶段流程。适合小红书、Instagram 等平台的轮播图设计。

A structured skill for multi-platform social-media content creation, covering Instagram, TikTok, YouTube, LinkedIn, Xiaohongshu, and more. The goal: outputs that satisfy each platform's nati 多平台社交媒体内容创作技能，覆盖 Instagram、TikTok、YouTube、LinkedIn、小红书等，产出符合各平台原生风格的图文内容。

A professional storyboard skill for film, advertising, short video, and educational narrative scenarios, built around a strict 'plan first, render later' flow. 面向影视、广告、短视频与教育叙事场景的专业分镜技能，采用严格的先规划、后渲染流程。适合分镜脚本与故事板设计。

Image matting tool: separates foreground from background and returns transparent background URL, suitable for product image processing, character cutout, and composition. 图像抠图工具：将前景与背景分离并返回透明背景的 URL，适用于产品图片处理、人物抠图和合成。

Convert static character images into vivid action videos with Jimeng Dream Actor. 使用即梦 (Jimeng) Dream Actor，将静态人物图片转化为生动的动作视频。

Generate dynamic videos based on a single first frame image and prompts using Jimeng. 使用即梦 (Jimeng) 首帧生视频模型，基于单张首帧图片和提示词生成动态视频。

Generate coherent transition videos using Jimeng's first and tail frame models. 使用即梦 (Jimeng) 首尾帧生视频模型，通过提供的第一帧和最后一帧图片生成连贯的过渡视频。

Generate realistic digital human broadcast videos from portrait images and audio/text using Jimeng OmniHuman 1.5. 使用即梦 (Jimeng) OmniHuman 1.5 模型，通过人像图片和音频/文本生成逼真的数字人播报视频。

Text-to-image generation with Jimeng, quickly converting text to high-quality images. 使用即梦 (Jimeng) 进行文生图创作，快速将文字转化为高质量图像。