openapi: 3.0.3 info: title: Fast MD description: | Convert any URL or HTML into clean, LLM-ready Markdown. Strips ads, nav, and clutter. Built for RAG pipelines and AI agents. version: 1.0.0 servers: - url: https://fast-md.onrender.com description: Production paths: /convert: post: summary: Convert HTML or URL to clean Markdown description: | Accepts either a `url` to fetch and convert, or raw `html` to convert directly. By default, extracts the main article content using smart heuristics — stripping nav, ads, sidebars, and footers. Set `extract_main: false` to convert the entire HTML document as-is. operationId: convertHtmlToMarkdown requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ConvertRequest' examples: url: summary: Convert from URL value: url: https://en.wikipedia.org/wiki/Markdown html: summary: Convert raw HTML value: html: "

Hello

Some bold text.

" urlNoExtract: summary: Full-page conversion (no main-content extraction) value: url: https://example.com extract_main: false responses: '200': description: Successful conversion content: application/json: schema: $ref: '#/components/schemas/ConvertResponse' example: markdown: "# Markdown\n\nMarkdown is a lightweight markup language..." tokens_estimated: 4823 characters: 19291 source_url: https://en.wikipedia.org/wiki/Markdown '400': description: Bad request — missing or invalid input content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' example: error: Provide either "url" or "html" in the request body. '500': description: Internal conversion error content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' /health: get: summary: Health check description: Returns OK if the service is running. operationId: healthCheck responses: '200': description: Service is healthy content: application/json: schema: type: object properties: status: type: string example: ok timestamp: type: string format: date-time components: schemas: ConvertRequest: type: object properties: url: type: string format: uri description: URL to fetch and convert. Provide this OR `html`. example: https://en.wikipedia.org/wiki/Markdown html: type: string description: Raw HTML string to convert. Provide this OR `url`. example: "

Hello

" extract_main: type: boolean default: true description: | If `true` (default), uses smart content extraction to grab only the main article body, stripping nav, ads, sidebars, footers. Set to `false` to convert the entire document. ConvertResponse: type: object required: [markdown, tokens_estimated, characters] properties: markdown: type: string description: The clean Markdown output. tokens_estimated: type: integer description: Approximate token count (1 token ≈ 4 chars). Use to budget LLM cost. characters: type: integer description: Character count of the markdown output. source_url: type: string nullable: true description: Echoes the input URL if one was provided. Null for raw HTML input. ErrorResponse: type: object required: [error] properties: error: type: string description: Human-readable error message. details: type: string description: Additional error context (when available).