Nano Gpt Mcp — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Nano Gpt Mcp (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
An MCP (Model Context Protocol) server for NanoGPT — giving you access to 700+ AI models for text, image, video, and audio generation directly from Claude Code or any MCP-compatible client.
NanoGPT is a pay-as-you-go AI gateway that provides access to models from OpenAI, Anthropic, Google, Meta, Mistral, DeepSeek, and many more — all through a single API key. No subscriptions required.
| Tool | Description |
|---|---|
nano_chat | Chat with 700+ text models (GPT, Claude, Gemini, Llama, DeepSeek, Qwen, Mistral, etc.) |
nano_generate_image | Generate images (GPT Image 1.5, Flux, Stable Diffusion, Recraft, HiDream, etc.) |
nano_generate_video | Generate videos (Kling, Sora 2, Veo 3, Wan, MiniMax, etc.) |
nano_list_models | Browse available models by type (text/image/video/audio) with pricing |
nano_check_balance | Check your account balance |
All responses include cost and remaining balance.
Sign up at nano-gpt.com and get your API key from the dashboard.
git clone https://github.com/mageaustralia/nano-gpt-mcp.git
cd nano-gpt-mcp
npm installAdd to your ~/.claude/settings.json to make it available globally (all projects):
{
"mcpServers": {
"nano-gpt": {
"command": "node",
"args": ["/path/to/nano-gpt-mcp/server.mjs"],
"env": {
"NANO_GPT_API_KEY": "your-api-key-here"
}
}
}
}Or add to a project-specific .claude/settings.json for per-project use.
The MCP server will be available after restarting Claude Code.
"Use nano_chat to ask GPT-4o what the best sorting algorithm is for nearly-sorted data"
"Ask DeepSeek R1 to solve this math problem using nano_chat"
"Use nano_chat with google/gemini-2.5-flash to review this code snippet"
"Use nano_generate_image with hidream-i1-fast to create a product photo of a laptop on a white background"
"Generate an image with gpt-image-1.5: a watercolor painting of a coastal town at sunset"
"Use nano-banana to generate a logo for a coffee shop called 'Bean There'"
"Use nano_generate_video with kling-v26-pro to create a 5-second video of ocean waves"
"Generate a video with wan-wavespeed-25: a timelapse of flowers blooming"
"Use nano_list_models to show me all image models"
"List video models on nano-gpt"
"Search for 'llama' in nano-gpt text models"
"Check my nano-gpt balance"
openai/gpt-4o, openai/gpt-4o-mini, openai/o3, openai/o4-minianthropic/claude-sonnet-4-5-20250929, anthropic/claude-opus-4google/gemini-2.5-flash, google/gemini-2.5-prodeepseek/deepseek-r1, deepseek/deepseek-v3meta-llama/llama-3.3-70b, meta-llama/llama-4-maverickqwen/qwen3-235b, mistralai/mistral-largegpt-image-1.5 - OpenAI's latesthidream-i1-fast - Fast high-quality generationflux-1.1-pro - Black Forest Labsrecraft-v3 - Design-focusedstable-diffusion-3.5-large - Stability AInano-banana - Budget-friendlykling-v26-pro - Text & image to video with audiosora-2 - OpenAI video generationveo3-video - Google Veo 3wan-wavespeed-25 - Fast generationminimax-hailuo-02-pro - High qualityThe MCP server translates tool calls into NanoGPT API requests:
/v1/chat/completions endpoint/v1/images/generations endpoint (returns URLs or base64)/v1/video/generations endpoint/v1/models, /v1/image-models, /v1/video-models, /v1/audio-modelsMIT
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.