skills.rest
opensquilla
opensquillaOFFICIAL
Explore all skills by this author →
WebsiteGitHub

short-drama-review-normalizer

Normalizes short-drama review replies into fail-closed consent decisions before paid media generation.

6.8k|540|Updated May 6, 2026
Search Tags:#media generation#fail-closed#short drama#consent gate#deterministic classification#bilingual parsing#review normalization
One-click install
npx skills add https://github.com/opensquilla/opensquilla --skill short-drama-review-normalizer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: short-drama-review-normalizer
Source: https://github.com/opensquilla/opensquilla/tree/main/src/opensquilla/skills/bundled/short-drama-review-normalizer
Command: npx skills add https://github.com/opensquilla/opensquilla --skill short-drama-review-normalizer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Free-form user replies to a short-drama draft review are ambiguous, and letting a language model interpret them risks triggering paid image or video generation without genuine consent. This Skill deterministically classifies review and confirmation replies so external provider calls only proceed on explicit approval.

Core Features & Use Cases

  • Deterministic Consent Gate: Classifies bilingual (English/Chinese) review replies into proceed, revise, hold, or cancel decisions using pattern matching, never a model or network call.
  • Privacy Restriction Detection: Holds generation when replies refuse external transfer, require on-device-only processing, or prohibit third-party access to content.
  • Canonical Script Snapshot: Freezes the final in-memory script (up to 200,000 UTF-8 bytes) bound to exactly one valid decision, without re-reading the user-editable script file.
  • Use Case: A user replies "change the style to anime, but do not generate yet" to a draft script; the Skill emits DECISION: hold with the adjustment preserved, and only a later standalone explicit approval unlocks the paid media step.

Quick Start

Pipe a JSON payload containing the review text and phase into the normalize script to receive a bounded decision block for the short-drama workflow.

Frequently Asked Questions about short-drama-review-normalizer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I gate paid AI media generation on explicit user consent?▼

Use a deterministic classifier that maps review replies to proceed, revise, hold, or cancel decisions before any provider call. This Skill emits a bounded DECISION block from pattern matching, so a language model can never promote an ambiguous reply into consent.

How to detect privacy restrictions in user review replies?▼

The normalizer detects negated transfer verbs, on-device or local-only confinement requirements, and prohibitions on third parties seeing or receiving content. Matching replies produce a hold decision with basis external_transfer_refused, even when the same reply contains a valid edit.

Does this consent gate support Chinese and English replies?▼

Yes, it includes parallel Chinese and English pattern sets for approvals, cancellations, generation deferrals, adjustments, and privacy restrictions. Replies are normalized for punctuation and tokenized per language before classification.

What happens when a review reply is ambiguous or off-topic?▼

Ambiguous, empty, questioning, or off-topic replies fail closed with DECISION: hold and a basis such as unclear_or_off_topic or revision_confirmation_required. Only explicit approval phrases or recognizable short-drama adjustments produce proceed or revise.

What are the limitations of regex-based consent classification?▼

Pattern matching only recognizes phrasings covered by its rule sets, so novel consent wording defaults to hold rather than proceed. It is intentionally fail-closed and deterministic, trading recall for a guarantee that unclear input never authorizes paid generation.

Related Skills

Explore More Software Engineering Skills→
tangffntrtangffntr
5

script-generator

Convert structured outlines into storyboard scripts with scenes and camera directions.

Community
Advanced
Arcanada-oneArcanada-one
11

dr-next-snapshot-replay

Resume staged tasks from validated snapshot files with a bilingual replay prompt.

Official
Advanced
sr-ai-devsr-ai-dev

mirror

Mirror requests back as What, Why, Scope, and Constraints with a confirmation question.

Official
Basic
jeremydhoover-blipjeremydhoover-blip
7

writing-confirmation-dialogs

Generate confirmation dialogs with risk-based UI treatments and character limits.

Community
Intermediate