io.github.farssaadeldin7-ux/emotional-resonance-analyzer

Open the Stripe billing portal to change plan, update the card, or cancel the Emotional Resonance Analyzer subscription.

Run the five drop-off tells mechanically against a scored scene table and Q&A ledger: tripped stretches by timecode and cause, ledger findings, overlaps with priority, and total exposed run-time. Threshold arithmetic on scores the editor supplies — it does not read the film, and a tripped tell is a stretch to watch again, not a proven defect. Requires a paid plan.

The five diagnosable drop-off causes — no question open, stakes not personalised, tonal monotony, premature resolution, texture starvation — each with its mechanical tell, threshold, standard fix and a worked example, plus the priority order when tells overlap. Omit cause to list all five.

The modelled engagement curve, drawn: one point per scene, computed deterministically from the method's existing arithmetic — the editor's intensity score, the valence movement between consecutive scenes, and a penalty per drop-off tell from the check_tells logic — with every flagged stretch shaded, written as a self-contained SVG line chart the editor can look at. Modelled from scene scores: it is not a measurement of any audience, predicts nothing, and a real retention curve outranks it entirely. Requires a paid plan.

The attention conventions of each encoded form — short_doc, feature_doc, branded, youtube_longform, broadcast: when the first question must open, typical departure points, the central-close window and the talking-head limit. Conventions as practitioners describe them, not audience measurements. Omit form to list all of them.

Store and activate a Emotional Resonance Analyzer licence key on this machine. Use when the user pastes a licence key, or after a purchase completes.

Check the Emotional Resonance Analyzer licence on this machine: plan, included capabilities, remaining quota and seat usage. Call this when a paid tool reports a licence problem, or when the user asks what their plan includes.

List the available Emotional Resonance Analyzer plans, prices and what each one includes.

Record an analysis in the local log — film, cut version, form, what was flagged, the three chosen cuts with effort ratings, and whether real retention data was supplied. Logging two versions of the same film is how a recut gets compared honestly. Requires a paid plan; the licence check is the only thing this tool sends anywhere.

The valence and intensity series read as a derivative: per-scene changes, flat stretches with the end scores quoted, the largest gap between valence changes of two or more points, intensity peaks and whether each one lands, and the last-quarter trend. Arithmetic on the editor's scores — flat stretches are the problem, not low stretches, and nothing here is a score of the film. Requires a paid plan.

Classify each drop in a real retention curve against the tripped tells: explained (lands on or just after a tell — with the cause), unexplained (usually performance, music or grade, which this cannot see), or model-only (a tell the audience never minded — the model was wrong there). The curve wins; no flag is adjusted to fit it. Requires a paid plan.

The local record of past analyses — list them, filter by film to set two versions of a cut side by side, or fetch one in full. Reading only; it makes no new claim. Requires a paid plan.

The scoring vocabulary the skill scores with: valence anchors (-3..+3), intensity anchors (0..5), the scene-boundary definition, the series-pattern readings, the Q&A ledger format and the edit effort scale. Scoring itself is a reading the editor or the skill makes — this only returns the anchors.

Begin a Emotional Resonance Analyzer purchase. Returns a Stripe Checkout link. Call list_plans first if the user has not chosen a plan.