ApilynDocument operations API
DevelopersOpen dashboard

Product

Every document stage, one API contract.

Apilyn separates orchestration (your app + our API) from execution (specialized worker pools). You choose pipelines; we enforce quotas, storage lifecycle, and webhook delivery semantics across pdf-core, ocr, office-core, web-core, and archive-core pools.

Developer guideOpen dashboard

Capability areas

Names below match public pipeline and tool identifiers. Enablement depends on your deployment profile—capability responses are explicit when a stage is unavailable.

Convert & merge

Office, image, and PDF pipelines with deterministic job IDs, signed downloads, and webhook completion events.

  • pdf.merge
  • pdf.split
  • office.convert
  • image.convert

OCR & structure

Extract text and layout from scans and photos. Route heavy OCR to dedicated worker pools with visibility timeouts tuned for large files.

  • ocr.extract
  • pdf.text_layer

PDF/UA & accessibility

Check and remediate accessibility profiles with explicit capability gates—no marketing claims beyond what your deployment enables.

  • pdf.accessibility_check
  • pdf.accessibility_remediate

PII & document intelligence

Detect and redact sensitive fields, generate embeddings, and keep tenant data isolated with org-scoped storage and RLS.

  • pii.detect
  • embeddings.index
  • search.semantic

Sync utilities

Low-latency calculators, validators, and formatters for control-plane adjacency—same API surface as async document jobs.

  • calculators.hash
  • validate.email
  • json.format

Web capture & archives

Headless capture, packaging, and archive operations isolated in web-core and archive-core pools.

  • web.screenshot
  • archive.create
  • archive.extract

Generated SDKs and the versioned OpenAPI document are the integration source of truth. Wire them into CI so production never drifts from the schema you tested.

Download OpenAPI

FAQ

Product questions

Which pipelines does Apilyn expose?

Public pipeline names include pdf.merge, pdf.split, office.convert, ocr.extract, pdf.accessibility_check, pdf.accessibility_remediate, pii.detect, embeddings.index, web.screenshot, archive.create, and sync tools such as calculators.hash and webhook.hmac_verify. Enablement depends on your deployment profile; capability responses are explicit when a stage is unavailable.

How are workers isolated?

Execution runs in named worker pools (pdf-core, ocr, office-core, web-core, and others) with prlimits and queue visibility tuned per pool. Your application orchestrates jobs; Apilyn enforces quotas, storage lifecycle, and webhook delivery semantics.

Do you ship SDKs?

Yes. Python and TypeScript SDKs are generated from the versioned OpenAPI JSON at api.apilyn.com/v1/openapi.json. Pin SDK versions in CI so error codes and field names match production.

How does storage and tenancy work?

Upload intents stream bytes to org-scoped object storage. Downloads use short-lived signed URLs. Tenant data is isolated with organization boundaries—treat staging and production as separate orgs, the same way you separate cloud accounts.