gmaxsoft/pdf-to-json-api

pdf-to-json-api is a Node.js REST service that downloads a PDF from a public URL and returns structured JSON extracted using Anthropic’s Claude model. It’s aimed at developers or backend services that need automated data extraction from documents (invoices, contracts, Polish KRS filings, or generic reports). Key features: - POST /extract: provide url plus built-in schema (krs, faktura, umowa, generic) or a customPrompt; optional model and includeRaw. - GET /schemas and GET /health endpoints. - Responses use {ok, data, meta} (includes source URL, model, pdf size, token usage); errors returned as {ok:false, error}. - Add custom schemas by editing src/schemas.ts. - Config via .env (ANTHROPIC_API_KEY required; PORT, MAX_PDF_SIZE_MB, REQUEST_TIMEOUT_MS available). - npm scripts for dev, build and start; examples provided for curl/Postman.

このRepoのdescription / GitHub topics / READMEのAI要約に出現したAI関連キーワードをカテゴリ別に表示。各バッジは該当カテゴリの詳細ランキングへリンクします。