Developers
Programmatic access to the Flaex AI directory: search hundreds of AI tools by intent, fetch any page as markdown, or connect agents natively over the Model Context Protocol.
Machine-readable OpenAPI 3.1 spec for every public endpoint, with operation IDs, typed schemas, and scoped security schemes.
https://www.flaex.ai/openapi.jsonGET /api/search?q=<query> — semantic search across the AI tools directory. Supports mode=semantic|hybrid|discover, pricing and category filters.
https://www.flaex.ai/api/search?q=ai+video+editor&limit=10POST JSON-RPC 2.0 to /api/mcp. Supports initialize, tools/list, and tools/call with tools flaex_search_tools and flaex_intent_discovery.
https://www.flaex.ai/api/mcpCapability document describing the Flaex MCP transport and available tools.
https://www.flaex.ai/.well-known/mcp/server-card.jsonStandard authorization-server metadata including supported scopes: search:read, directory:read, newsletter:write, mcp:invoke.
https://www.flaex.ai/.well-known/oauth-authorization-serverRFC 9391-style linkset catalog of public API resources.
https://www.flaex.ai/.well-known/api-catalogMarkdown guide for AI agents: key sections, data formats, and conventions.
https://www.flaex.ai/llms.txtThe public API supports scoped permissions so callers only request the access they need. Scopes are declared in the OpenAPI security schemes and the OAuth metadata.
search:read — read tool search resultsdirectory:read — read directory listings and metadatanewsletter:write — manage newsletter subscriptionsmcp:invoke — call Flaex MCP toolsBuilding an integration? Start with the llms.txt agent guide, then explore the Flaex AI directory.