Small decisions. Interesting possibilities.Submit contentSubmit
Agent tooling

A page-by-page OCR router

Doc-OCR router using Jev @typesafeai A Jev-powered router that looks at a PDF page by page, decides which pages actually need OCR, extracts the rest locally. Result: save cost on # OCR pages + speed

Added to Jevfast
X post ↗

How it uses Jev

Jev decides page by page which PDF pages need OCR; the rest are extracted locally.

App workflow

  1. Inspect a page

    The router considers PDF pages individually.

  2. Decide whether OCR is needed

    Jev determines which pages need OCR.

  3. Extract content

    Pages selected for OCR use OCR; the rest are extracted locally.

Why it is interesting

Page-level OCR routing can reserve expensive OCR for pages that need it while keeping text extraction local on the rest. The decision acts on each page rather than applying one rule to the whole PDF.

  • Jot

    A tool-running agent lets Jev choose actions and arguments from the current conversation.

  • Jev Use

    Give agents typed Jev decisions with explicit screening and escalation to their host.

  • Nerve for Hermes

    Supervise Hermes agent work with Jev decisions, evidence checks and explicit completion authority.

Explore more in Agent tooling →