Which agent runtimes do skills support?

Skills work with Claude Code, Codex, Gemini CLI, Cursor, Windsurf, Roo Code, OpenClaw, Hermes Agent, and 50+ more agent runtimes that support the skills protocol.

Do I need a separate account per provider?

No. A single RunAPI API key unlocks all 26 providers. You do not need separate accounts with Anthropic, OpenAI, Google, or any other provider.

Yes. Skills are free and open-source. API usage is billed through your RunAPI account with pay-as-you-go pricing. No subscriptions or minimum spend. Failed generations are not charged.

Where do I report a broken skill?

Open an issue on the skill's GitHub repository at github.com/runapi-ai/ , or contact contact@runapi.ai.

Can I use skills alongside the MCP server and CLI?

Yes. Skills, the MCP server, and the CLI all share the same RunAPI API key and account. A skill installed in Claude Code can call the same models you access through the MCP server or CLI. Tasks appear in the same dashboard regardless of which interface created them.

Claude visitors: new accounts receive $1.00 in bonus API credit within 30 days. One per person; usage credits only.

Agent Skills

Teachable Skills for AI Agents

Give your AI coding agent domain-specific capabilities. Install a skill, and your agent learns how to generate video, create music, edit images, and more — all through RunAPI.

Browse Skills Build Your Own

Works with Claude Code · Codex · Gemini CLI

Install

# Install a skill for Claude Code
claude skill add runapi/video-generation

# Your agent can now generate videos
"Generate a 5-second video of a sunset
 using Kling v2.1"

✓ Skill installed: video-generation
  3 tools available

50+

Skills

Modalities

Code required

MCP

Protocol

Skills for Every Modality

VIDEO

Video Generation

Generate and manage video tasks with Kling, Seedance, Veo, and more.

IMAGE

Image Generation

Create, edit, and upscale images with Flux, Midjourney, and GPT Image.

MUSIC

Music Generation

Compose full tracks with vocals using Suno and Udio.

AUDIO

Audio Processing

Text-to-speech with ElevenLabs, transcription with Whisper.

LLM

LLM Gateway

Chat completions via Claude, GPT, Gemini with one unified schema.

UTILITY

Model Catalog

Browse models, check pricing, and manage credits from your agent.

Browse Skills

10 Skills

Claude

Anthropic

Claude API access for Anthropic's LLM across complex reasoning, code, analysis, and extended-context tasks.

npx skills add runapi-ai/claude -g

View →

DeepSeek

DeepSeek API access via RunAPI — flash for fast, low-cost work; pro for complex agentic tasks.

npx skills add runapi-ai/deepseek -g

View →

Gemini

Google

Gemini API access for Google's multimodal LLM across chat, code generation, reasoning, and long-context tasks.

npx skills add runapi-ai/gemini -g

View →

GLM

Z.ai

Z.ai GLM API access via RunAPI — MIT-licensed MoE models with up to 200K context, leading open-weight coding benchmarks.

npx skills add runapi-ai/glm -g

View →

GPT

OpenAI

GPT API access for OpenAI's flagship LLM across chat, code generation, and multi-step reasoning tasks.

npx skills add runapi-ai/gpt -g

View →

Grok

xAI

xAI Grok 4.5 and 4.6 API access through RunAPI Responses and Chat Completions for reasoning, tools, and structured output.

npx skills add runapi-ai/grok -g

View →

Kimi

Moonshot AI

Moonshot AI Kimi API access via RunAPI — kimi-k3 is the current flagship with always-on reasoning; kimi-k2.5 and kimi-k2.6 remain available.

npx skills add runapi-ai/kimi -g

View →

MiMo

Xiaomi

MiMo API via RunAPI — base mimo-v2.5 supports text and remote HTTP(S) images in synchronous OpenAI Chat Completions; mimo-v2.5-pro remains text-only.

npx skills add runapi-ai/mimo -g

View →

MiniMax

MiniMax text API access via RunAPI — sparse MoE models from 200K to 1M context, up to 80.5% SWE-bench Verified.

npx skills add runapi-ai/minimax -g

View →

Qwen

Alibaba

Qwen API access via RunAPI: open-weight Qwen3.5, Qwen3.6, Qwen3-VL and Qwen3-Coder-Next language models on OpenAI-compatible Chat Completions.

npx skills add runapi-ai/qwen -g

View →

HOW IT WORKS

From Install to First Generation

Install

One command to add a skill. Your agent discovers its tools automatically.
Describe

Tell your agent what to create in natural language. The skill handles parameters and API calls.
Generate

Results are saved locally or streamed back. Costs are shown before and after.