Scoring & Critique
Five tools for evaluating video quality. Score candidates on six dimensions, check brand finish per personality, audit product demo clarity, evaluate frame strips for visual quality, and detect motion density issues.
score_candidate_video
The primary scoring tool. Evaluates a candidate video across 6 weighted dimensions: hook, narrative_arc, clarity, visual_hierarchy, motion_quality, and brand_finish. Produces per-scene subscores for targeted revision.
- Name
manifest- Type
- object
- Description
Sequence manifest to score.
- Name
brief- Type
- object
- Description
Original brief to evaluate adherence against.
Returns: Overall score (0-100), per-dimension scores with weights, per-scene subscores, and improvement notes.
Try asking your AI:
"Score this candidate video on all six dimensions"
score_brand_finish
Evaluate compositing quality per personality and style. Checks that the visual treatment matches the intended brand feel, including color grading, grain, typography, and surface treatments.
- Name
manifest- Type
- object
- Description
Sequence manifest to evaluate.
- Name
personality- Type
- string
- Description
Target personality to check brand finish against.
- Name
style- Type
- string
- Description
Style pack to check adherence.
Returns: Brand finish score (0-100), per-element quality notes, and suggestions for improving brand consistency.
Try asking your AI:
"Score the brand finish for this prestige-style editorial sequence"
score_product_demo_clarity
Evaluate product interaction truthfulness, camera intent alignment, pacing, and visual hierarchy for product demo scenes. Catches common issues like unclear UI interactions, mismatched camera moves, and rushed pacing.
- Name
scenes- Type
- object[]
- Description
Product demo scenes to evaluate.
Returns: Clarity score (0-100), per-scene assessments for interaction truth, camera intent, pacing, and hierarchy.
Try asking your AI:
"Score the product demo clarity for my SaaS walkthrough scenes"
score_frame_strip
Visual quality evaluation for a strip of frames. Checks contrast, readability, hierarchy, brand consistency, and pacing rhythm across a sequence of still frames.
- Name
frames- Type
- object[]
- Description
Array of frame objects (from contact sheet or captures).
Returns: Per-frame quality scores, cross-frame consistency analysis, and visual rhythm assessment.
Try asking your AI:
"Score this frame strip for visual quality and rhythm"
audit_motion_density
Audit motion density across a sequence and suggest simplifications. Detects scenes where too many elements are animating simultaneously, competing for attention, or exceeding comfortable motion budgets.
- Name
manifest- Type
- object
- Description
Sequence manifest to audit.
Returns: Per-scene motion density scores, flagged overloaded scenes, and specific simplification suggestions.
Try asking your AI:
"Audit motion density and suggest where to simplify"