Atlas Cloud Skills

🎬 Seedance 2.5 is now live on Atlas Cloud! ByteDance's newest flagship video model — native audio-video joint generation, up to 30s in a single generation, native 480p/720p/1080p (up to 4K via built-in super-resolution), aspect ratios from 21:9 to 9:16, and multimodal reference of up to 30 images + 10 videos + 10 audio clips. Available now: Text-to-Video · Image-to-Video · Reference-to-Video from $0.134/s. Cheaper tiers: Seedance 2.0 Fast from $0.072/s · Seedance 2.0 Mini from $0.039/s.

🔓 Need the full-power build? The unrestricted / full-capability pipeline — fewer guardrails, broader subject range, max-fidelity output — is available through Atlas Cloud Workflow. Hook it straight into your skill via the same API key.

🎬 Newest video generations — Kling 4.0 · Wan 3.0. The latest Kling and Wan alongside Kling 3.0 / Video O3 and Wan 2.7 — same API key, same endpoints, browse the full video line-up.

Use Atlas Cloud's 300+ image / video / LLM models inside Claude Code, Codex, Gemini CLI, and other AI coding agents. Generate images, videos & chat via curated Skills.

→ Get your free Atlas Cloud API key — 300+ models, one key, OpenAI-compatible.

Supported Models

• 🎬 Video (200) — MiniMax H3 Max · MiniMax H3 Fast · Gemini Omni 1.1 Flash · MiniMax H3 · Wan-3.0-Prime · Wan-3.0

• 🎨 Image (130) — GPT Image 2.5 Sunburst · GPT Image 2.5 Flare · Seedream v4.7 · MAI-Image-2.5-Pro

• 🧊 3D (7) — Seed3D 2.0 · Tripo H3.1 · Hunyuan 3D Rapid · Hunyuan 3D Pro

• 💬 LLM (70) — DeepSeek V4 Pro 0813 · Grok 4.6 · DeepSeek V4 Flash Vision Exp · DeepSeek V4 Flash 0731

• 🔊 Audio (TTS · Music · ASR) (13) — Seed Audio 1.0 · MiniMax Speech 2.6 Turbo · MiniMax Speech 2.6 HD · xAI TTS v1

• 📚 Explore more — all 443 live models »

Contents

• Supported Models
• Featured Recipes
• Available Skills
  • atlas-cloud
  • seedance-2-5-skill
  • universal-video-prompt-skill
• Installation
• Setup
• What You Can Do
• MCP Server
• More Atlas Cloud Tools

Featured Recipes

Don't start from a blank prompt — start from a workflow. Three to try first:

• 🛍️ Product Render → 30-Second Ad — turn a product still into a short ad clip
• 🎭 Character Sheet → Multi-Shot AI Drama — one character, consistent across shots
• 📱 Long Video → Vertical Short — reframe + animate for TikTok / Reels / Shorts

Browse all 25 recipes in the library » — across 🎨 visual · 🎬 motion · ✂️ edit · 📱 social.

Available Skills

atlas-cloud

Quickly integrate Atlas Cloud API into your projects. This skill provides:

• Complete API reference for image generation, video generation, LLM chat, media upload, and quick generation
• All 9 MCP tools documented: atlas_list_models, atlas_search_docs, atlas_get_model_info, atlas_generate_image, atlas_generate_video, atlas_quick_generate, atlas_chat, atlas_get_prediction, atlas_upload_media
• Ready-to-use code templates in Python, Node.js/TypeScript, and cURL
• Popular model IDs with pricing info
• OpenAI SDK compatibility guide for LLM models
• Error handling, retry strategy, and best practices

seedance-2-5-skill

A model-specific sub-skill for controllable Seedance video, layered on top of atlas-cloud. Use it when a job needs shot planning rather than a single prompt:

• Route selection first — text-to-video, storyboard-image-to-video, asset-reference-to-video, first-and-last-frame, or extending an existing clip
• Reference discipline — how to assign people / product / scene / style / audio references so identity holds across shots
• Shot craft — camera, lighting and composition vocabulary, transition patterns, long-video continuity, real-person handling
• Editing and extension — rewriting part of an existing video, changing lighting or style, extending a short clip
• Execution — submits through Atlas Cloud (MCP, CLI or REST) and verifies model availability before a billable run
• Every reference ships in English and Simplified Chinese; Chinese requests follow a dedicated Chinese workflow

Works with universal-video-prompt-skill when the same brief needs to run on more than one model. Prompt library for this model: awesome-seedance-2.5-prompts-skills

universal-video-prompt-skill

Model-agnostic companion to seedance-2-5-skill. It writes one prompt spec — scope, locks, staging, end states — separately from the dialect that expresses it, then compiles that spec for whichever video model you can actually call. Use it when:

• The same brief has to run on several models, or the target model is not available yet and the work must proceed elsewhere
• You are building a model-comparison matrix and need the prompts to differ only by dialect
• A prompt must survive a model swap instead of being rewritten

Each model carries a measured profile (reference syntax, limits, timing adherence, default-bias behaviour); the skill probes what it does not know, degrades the spec to what the model supports, and reports every degrade. English + Simplified Chinese references throughout.

Installation

One-Line Install

npx skills add AtlasCloudAI/atlas-cloud-skills

Shell Script

curl -fsSL https://raw.githubusercontent.com/AtlasCloudAI/atlas-cloud-skills/main/install.sh | sh

Manual

Copy atlas-cloud/ to ~/.claude/skills/atlas-cloud/, and any sub-skill under skills/ to ~/.claude/skills/<name>/.

Install one skill only

curl -fsSL https://raw.githubusercontent.com/AtlasCloudAI/atlas-cloud-skills/main/install.sh | sh -s atlas-cloud
curl -fsSL https://raw.githubusercontent.com/AtlasCloudAI/atlas-cloud-skills/main/install.sh | sh -s seedance-2-5-skill
curl -fsSL https://raw.githubusercontent.com/AtlasCloudAI/atlas-cloud-skills/main/install.sh | sh -s universal-video-prompt-skill

seedance-2-5-skill references universal-video-prompt-skill, so install both if you want cross-model prompt specs.

Setup

1. Get an API Key at Atlas Cloud Console
2. Set the environment variable:

export ATLASCLOUD_API_KEY="your-api-key-here"

See .env.example for a ready-to-copy template.

What You Can Do

| Capability | Endpoint | Example Models | |------------|----------|----------------| | Image Generation | POST /api/v1/model/generateImage | Nano Banana 2, Seedream v5.0, Z-Image | | Video Generation | POST /api/v1/model/generateVideo | Seedance 2.5, Kling 4.0, Wan 3.0, Kling Video O3 | | Audio — TTS & Music | POST /api/v1/model/generateAudio | Seed Audio 1.0, Suno Chirp v5, MiniMax Music | | Speech-to-Text (ASR) | POST /api/v1/model/generateAudio | Seed ASR 2.0, xAI STT | | 3D Generation | POST /api/v1/model/generateImage | Seed3D 2.0, Hunyuan 3D (image/text-to-3D) | | LLM Chat | POST /v1/chat/completions | Qwen3.5, Kimi K2.5, DeepSeek V3.2, GLM 5 | | Upload Media | POST /api/v1/model/uploadMedia | Upload local files to get public URLs | | Quick Generate | Auto model search + submit | One-step generation by keyword | | Search Models | Fuzzy search by keyword | Find models by name, type, or provider |

MCP Server

For a more native experience, install the Atlas Cloud MCP Server:

CLI Tools (One-Line Install)

# Claude Code
claude mcp add atlascloud -- npx -y atlascloud-mcp

# Gemini CLI
gemini mcp add atlascloud -- npx -y atlascloud-mcp

# OpenAI Codex CLI
codex mcp add atlascloud -- npx -y atlascloud-mcp

IDEs & Editors (JSON Config)

{
  "mcpServers": {
    "atlascloud": {
      "command": "npx",
      "args": ["-y", "atlascloud-mcp"],
      "env": {
        "ATLASCLOUD_API_KEY": "your-api-key-here"
      }
    }
  }
}

Supports Cursor, Windsurf, VS Code (Copilot), Trae, Zed, JetBrains, Claude Desktop, ChatGPT Desktop, Amazon Q Developer, Cline, Roo Code, Continue, and all MCP-compatible clients.

More Atlas Cloud Tools

• 🧰 Want to use it from the terminal? → atlascloud-cli
• 🤖 Want to use it in Claude Code / Cursor? → Atlas Cloud MCP Server
• 🎬 Want it as a Claude Code / Codex / Gemini CLI Skill? → atlas-cloud-skills
• 🎨 ComfyUI nodes → atlascloud_comfyui
• 🔁 n8n nodes → n8n-nodes-atlascloud
• 💬 Join our Discord → discord.gg/MWmMr4q9es
• 🌐 Website → atlascloud.ai

Docs: Models & API · Console / API Keys