still-qa

Installation
SKILL.md

still-qa

python scripts/judge.py --endpoint $VLM_ENDPOINT --model $VLM_MODEL --image out/s03.png --prompt "<shot prompt>" --bible bible.md [--ref cast/mara_front.png]

This prints JSON with scores, mean, verdict (pass/retry), issues and fix.

The loop

  1. Generate the still, then judge it.
  2. If the verdict is pass (every score 3 or higher and a mean of 3.5 or higher), accept it.
  3. If the verdict is retry, apply the single fix: a prompt edit, a new seed, or an image-edit pass. Regenerate and judge again. Allow at most 2 retries, then show the best take to the user with its issues.
  4. Log every take's scores in the plan so continuity-check can use them.

Judge setup

  • Serve an open-weights VLM (Qwen2.5-VL-7B-Instruct, Apache-2.0, unverified for your build) with vLLM, llama.cpp server, Ollama or LM Studio. Any OpenAI-compatible /v1/chat/completions endpoint works. A 7B VLM needs about 16 GB in bf16, or about 6-8 GB quantized (estimate, unverified).
  • In Comfyrizz-style setups, the orchestrator's own judge (for example Pi) can replace judge.py as long as it keeps the same rubric and JSON shape.
  • The judge must see the prompt and the bible descriptors, not only the image.
  • For video, judge 3 frames: the first, the middle and the last.
Installs
75
GitHub Stars
49
First Seen
Nov 1, 2026