FetchHarbor

Endpoints (12 live)

• GET /html-to-md — Convert a supplied HTML string into deterministic, LLM-ready Markdown. Preserves headings, links, lists, emphasis, and code without fetching a URL or invoking an AI model. Input is bounded to 2,000,000 characters. (0.005 USDC on Base)
• POST /chat — Generate one bounded assistant response with the operator's self-hosted Ollama model. Use for short, single-message inference where local processing is preferred. Accepts up to 8,000 characters and returns the model name, response text, and token counts when Ollama supplies them. (0.01 USDC on Base)
• GET /scrape — Fetch one public HTTP(S) URL for research, extraction, or agent workflows. Returns the final URL, upstream status, content type, and decoded content. Private, local, metadata, and reserved destinations are blocked; redirects are revalidated and response size and duration are bounded. (0.01 USDC on Base)
• POST /v1/audio/speech — Generate private, locally processed English speech from bounded text. Returns WAV audio as base64 JSON, uses a fixed voice allowlist, and does not retain or log submitted text. (0.04 USDC on Base)
• POST /audio/transcribe-summary — Transcribe bounded audio and explicitly pass the resulting text to the operator's local Ollama model for a concise summary. No input or output content is retained. (0.05 USDC on Base)
• POST /audio/transcribe — Transcribe up to five minutes of base64 audio in the isolated local media worker. Returns detected language, duration, and text without retaining the source audio or transcript. (0.025 USDC on Base)
• POST /audio/convert — Convert bounded audio to WAV, MP3, Opus, or FLAC with fixed server-side FFmpeg arguments in an isolated worker. Returns base64 JSON and retains no media. (0.015 USDC on Base)
• POST /audio/subtitles — Transcribe bounded audio and generate deterministic SRT or WebVTT subtitle text locally. Audio, transcripts, and subtitles are not retained. (0.035 USDC on Base)
• POST /pdf-parse — Extract embedded text from a public PDF URL or multipart upload for research and retrieval pipelines. Returns text, page count, and character count; scanned-image OCR is not included. Remote and uploaded PDFs are limited to the operator-configured maximum size. (0.01 USDC on Base)
• POST /html-to-md — Convert a supplied HTML string into deterministic, LLM-ready Markdown. Preserves headings, links, lists, emphasis, and code without fetching a URL or invoking an AI model. Input is bounded to 2,000,000 characters. (0.005 USDC on Base)
• POST /scrape — Fetch one public HTTP(S) URL for research, extraction, or agent workflows. Returns the final URL, upstream status, content type, and decoded content. Private, local, metadata, and reserved destinations are blocked; redirects are revalidated and response size and duration are bounded. (0.01 USDC on Base)
• GET /pdf-parse — Extract embedded text from a public PDF URL or multipart upload for research and retrieval pipelines. Returns text, page count, and character count; scanned-image OCR is not included. Remote and uploaded PDFs are limited to the operator-configured maximum size. (0.01 USDC on Base)