Video2Guide / tests /test_llm_extract.py

Commit History

  1. Add AI slide-to-transcript alignment for untimestamped images⁠

    Claude +1: Claude Opus 4.8
    e646192
  2. fix: images-mode ZIP dropped all images; retry malformed LLM JSON⁠

    Claude
    f09a075
  3. fix: parse LLM output ourselves so it works through the Parley proxy⁠

    Claude
    e5c847b
  4. feat: route MIT Parley keys to the Parley proxy (alongside normal Anthropic keys)⁠

    Claude
    897f74e

  1. fix: close the 3 previously-unverified low-severity audit findings⁠

    Claude
    0658e45
  2. fix: flatten RGBA/palette PNGs before JPEG-encoding for the LLM⁠

    Claude
    13e0c0c

  1. editor: remove summary / key-terms / self-check (round 3)⁠

    Claude
    b863381
  2. editor: long-form image description field (round 2)⁠

    Claude
    cf53b59

  1. llm_extract: cut per-segment API cost and add transient-error retry⁠

    Claude
    3602e87