Evaluate rendered UGC videos and generate eval.json and eval-report.md.
npx skills add https://github.com/alecs5am/ralphy --skill evaluator-alecs5am
Please help me install this Agent Skill. Skill: evaluator Source: https://github.com/alecs5am/ralphy/tree/main/.agents/skills/evaluator Command: npx skills add https://github.com/alecs5am/ralphy --skill evaluator-alecs5am
Quickly assess the quality and integrity of rendered UGC videos by producing a structured evaluation report that highlights scene segmentation, audio, captions, and visual consistency. The fixer agent can read the report to apply targeted fixes.
Run the evaluator on a rendered mp4 using ralphy eval video <path-to-mp4>, then review the generated eval.json and eval-report.md for the results.
High-intent search queries and answers about installing and using this skill.
Automate video quality evaluation by running the evaluator on a rendered mp4 to generate an eval.json and eval-report.md highlighting scene segmentation, audio, captions, and visual consistency.
Yes, you can run mid-render quality checks on UGC videos. The evaluator assesses quality across structure, keyframe, native-video, and deep-style render modes for both final cuts and intermediate renders.
Yes, the video evaluation report supports downstream fixer agents by outputting deterministic findings and tailored handoffs in a machine-readable eval.json contract for targeted automated fixes.
You can improve UGC scene analysis accuracy by optionally incorporating project context from scenario.json, BRIEF.md, and STYLE_LOCK.md files when evaluating rendered video outputs.
The best way to generate a machine-readable eval.json contract for artifact detection is using an automated evaluation tool that analyzes rendered video outputs and surfaces actionable findings.
GarrettRoiGenerate colored ASCII character videos from video, audio, images, or generative parameters.
alecs5amScore a project against its workspace rubric and generate a per-criterion report.
LeandelosantosCreate Remotion video compositions from React code with frame-driven animations.
gqy20Analyze sampled Manim video frames to detect visual artifacts and blocking issues.
