auditar-reel

Audits Reel scripts and captions against a 27-item quality gate with six mechanical check dimensions.

Updated Jul 13, 2026
One-click install
npx skills add https://github.com/oleonardorosso/reels-com-claude-code --skill auditar-reel-oleonardorosso
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: auditar-reel
Source: https://github.com/oleonardorosso/reels-com-claude-code/tree/main/.claude/skills/auditar-reel
Command: npx skills add https://github.com/oleonardorosso/reels-com-claude-code --skill auditar-reel-oleonardorosso

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Writing short-form video scripts that follow a strict editorial framework is error-prone: banned vocabulary slips in, the niche reveal happens too early, statistics lack sources, and calls to action repeat. This Skill mechanically audits a Reel script (spoken script plus Instagram caption) against a 27-item exit gate before approval or recording, returning a scored report with exact violations and suggested fixes. ## Core Features & Use Cases - Six audit dimensions: banned vocabulary (zero-tolerance greps), hidden-cause and trigger-word frequency control, silent funneling with reporter-only voice, Big Idea verification in movement 5, source coverage for every statistic, and emotional arc plus retention devices. - 27-item mechanical gate: each item runs as a concrete check (grep patterns, word-position ratios, timing coverage, hook family diversity) with PASS / AJUSTAR / NO-GO verdicts. - Batch scoreboard: when auditing a full weekly batch, it checks naming-form rotation, hook family diversity across Reels, and CTA cadence. - Use Case: Before recording a weekly batch of 5 Reels, run the audit on the batch file to catch a niche term leaking into the hook, a repeated trigger word, or an unsourced statistic, then apply the suggested fixes. ## Quick Start Ask the assistant to audit the Reel script in your output folder against the exit gate and report the score and verdict before recording.

Frequently Asked Questions about auditar-reel

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

FAQPage Schema
How do I audit a Reel script before recording?▼

Provide the path to the script file containing the 7 spoken movements plus the Instagram caption. The audit runs six check dimensions and the 27-item gate, then returns a score out of 100 with a GO, AJUSTAR, or NO-GO verdict and line-level fixes.

What does the Reel script quality gate check?▼

It checks banned vocabulary and em-dashes, niche terms leaking before the 60% mark, trigger-word repetition, hidden-cause mentions, source coverage for every statistic, emotional arc, retention devices, CTA uniqueness, and hook family diversity across a batch.

Can I audit a full weekly batch of Reels at once?▼

Yes. Batch mode audits each Reel individually and adds batch-level checks: naming-form rotation across Reels, at least 3 distinct hook families, and CTA cadence balance between conversion and engagement.

Why does the audit stop before running any checks?▼

The audit halts if required persona fields in the dados/ configuration files are still unfilled placeholders. You must complete the persona setup interview first so the vocabulary, trigger-word, and positioning lists can be assembled at runtime.

What causes an automatic NO-GO verdict?▼

Any zero-tolerance failure triggers NO-GO: an em-dash anywhere, banned spoken formulas, the niche bridge appearing before 55% of the script, the hidden cause being named, undefended absolutes, or an argument chain step without a supporting spoken line.