Developer API

Free privacy-first dev tools — plus a no-account API for the jobs a browser can't do. Pay per call with a credit key you store like an API key. No signup, no account, no subscription.

Available now

• HTML / PDF / DOCX → clean Markdown — at scale, server-side, for pipelines and RAG ingestion. 1 credit / page.
• AI OCR — text out of scanned images (and image-only PDFs). 3 credits / page.
• Document → typed JSON — PDF, HTML, email (.eml), and text into JSON matching your schema (invoices, receipts, emails → fields). 5 credits / call. The free PDF to Markdown & JSON tool is the in-browser preview.
• Domain / cert intelligence — a domain's Certificate-Transparency subdomain footprint + lookalike / typosquat candidates, each DNS-resolved and risk-scored. 2 credits / call.

How the no-signup model works

You buy a credit key (a random bearer token, e.g. bfk_…) with a one-off Stripe payment — packs from $10. The key is your balance: no email, no account, no password. Send it as the x-credit-key header on each request; calls debit credits until the key runs out, then top it up. Lose the key, lose the balance — store it like any API key (or use the emailed recovery link).

Endpoints

Document → Markdown

curl https://bytefork.tools/api/extract \
  -H "x-credit-key: bfk_your_key_here" \
  -F "[email protected]" \
  -F "format=md"

Returns Markdown. format=json returns a JSON envelope (markdown + page count + mode). Image files are OCR'd automatically.

Document → typed JSON (your schema)

curl https://bytefork.tools/api/extract \
  -H "x-credit-key: bfk_your_key_here" \
  -F "[email protected]" \
  -F 'schema={"type":"object","properties":{"vendor":{"type":"string"},"total":{"type":"number"},"currency":{"type":"string"}},"required":["vendor","total"]}'

Returns { data: { … } } matching your schema. Works on PDF, HTML, DOCX, images, and raw .eml email.

Domain / cert intelligence

curl "https://bytefork.tools/api/domain-intel?domain=example.com" \
  -H "x-credit-key: bfk_your_key_here"

Returns the CT subdomain list + lookalike domains with registered + risk (high/medium/low) for each.

Pricing

Pay-per-call, no monthly minimum. 1 credit ≈ 1 text page of a document; image/OCR pages cost 3; a typed-JSON extraction is 5 credits/call; a domain-intel lookup is 2. Packs:

• $10 → 1,500 credits (~$0.0067/page)
• $25 → 5,000 credits (~$0.005/page)
• $60 → 15,000 credits (~$0.004/page)

Exact pack sizes on the pricing page. You only ever spend what's on your key — nothing recurring.

Why no signup

The whole site is private by default and account-free. The API keeps that promise: a credit key is a wallet, not an identity. It's also the right shape for the real buyer — a program that wants a clean endpoint and per-call billing, not a dashboard login.

Questions or volume needs? Email us.