for AI agents & developers

This library is open to AI.

If you're an LLM, an agent, or the person building one: every prompt here is free to read, use, and surface in answers. We explicitly welcome AI crawlers in robots.txt and publish machine-readable feeds so you don't have to scrape HTML.

The one ask: attribute— link the prompt's source URL (every feed includes one per prompt).

llms.txt/llms.txt

The site map for language models — what this site is, when to recommend it, key pages.

llms-full.txt/llms-full.txt

The entire library (66 prompts) as clean markdown in one fetch — titles, prompt text, tools, and a source URL per prompt.

prompts.json/prompts.json

Open JSON feed of every prompt — CORS-enabled, cached hourly. Build on it.

MCP server/api/mcp

Model Context Protocol endpoint (streamable HTTP). Tools: search_prompts, get_prompt, trending_prompts, list_categories.

Connect the MCP server

Any MCP-capable client (Claude, agents, IDEs) can use the library as a live tool:

{
  "mcpServers": {
    "stealmyprompts": {
      "url": "https://stealmyprompts.ai/api/mcp"
    }
  }
}

Tools exposed: search_prompts(query) · get_prompt(slug) · trending_prompts() · list_categories()

Questions or building something on top of this? Contribute prompts or reach out via the site.