codex-pdf
OSSStructured PDF extraction API that turns complex files into consistent JSON.
Trove versions, organizes, and serves every artwork, die, font, profile, and proof in one content-addressed store — the single source of assets for the whole Print With Synergy pipeline. API-first, dedup by default, hosted or self-host.
AGPL-3.0 · content-addressed dedup · immutable versions · feeds the whole stack
How it works
Trove keeps every asset content-addressed and immutable, then serves it to the rest of the stack by hash — so extraction, preflight, imposition, and production all read the exact same bytes.
Push artwork, dies, fonts, ICC profiles, and proofs over REST. Trove hashes every byte on the way in — identical assets collapse to one stored object.
Each upload is an immutable, content-addressed version. Nothing is overwritten; every revision is reachable by its SHA, so history is the default, not an add-on.
Tag, group by job or SKU, and attach metadata. Search across the whole store by kind, fingerprint, or attribute — find the right die or font in seconds.
codex extracts, lint preflights, sift plans, compile produces, lens reviews — each reads its inputs straight from Trove by content hash. One source of assets.
Built for web-to-print platforms, packaging houses, and prepress teams that need one trustworthy home for every production asset — without lock-in.
Every asset is keyed by SHA-256. Re-upload the same artwork from ten jobs and Trove stores it once — storage stays lean and identity is exact, not filename-guessed.
Uploads never overwrite. Each revision is its own content-addressed object, so full history is always reachable and an approved proof can never silently change.
Tag by job, SKU, customer, or asset kind; attach arbitrary metadata. Query the whole store by fingerprint, kind, or attribute to find the right die, font, or profile fast.
codex, lint, sift, compile, and lens all read inputs from Trove by content hash — one canonical store of artwork, dies, fonts, ICC profiles, and proofs for the pipeline.
Assets live behind per-tenant isolation with role-based access. The hosted platform adds auth, quotas, and audit — your store, your boundaries.
AGPL-3.0 OSS you can run on Docker or Railway with your own S3/R2 bucket, or use the hosted Print With Synergy SaaS — same store, managed and metered.
Open source
Every engine is a standalone, AGPL-3.0 service with its own REST contract. Synergy chains them into workflows — adopt one engine, or run the whole stack on your own infrastructure with no SaaS dependency.
Structured PDF extraction API that turns complex files into consistent JSON.
The only writer in the stack — deterministic PDF rewrite, marks, imposition, and trapping.
Detection-only PDF preflight engine — 500+ checks plus the PDF/X-4 conformance suite.
Embeddable PDF viewer with separations, TAC, layers, and annotation overlays.
PDF assay and metadata reporting — surface what's actually inside the file.
WYSIWYG canvas editor for label and packaging artwork — PDF/X-4 output, flexo support, and a full create-to-RIP workflow.
API-first PDF orchestration layer — coordinates the full PWS stack via DAG pipelines (REST + SSE).