技能 listening-script

📦

listening-script

Name: listening-script
Author: godgod126

內容修訂 r1 安全

Create Listening Scripts That Preserve Evidence

Screen-oriented articles and reports can be difficult to follow aloud without losing important evidence. This skill adapts them into narration with clear comparisons, audible uncertainty, and separate source notes.

支援: Claude Codex Code(CC)

📊 74 尚可

Maintain this Skill? Claim it 📊 74 尚可

使用我的智慧代理安裝

將此請求複製給你的智慧代理，其中包含權威 Skill 頁面和清單。

智能體請求

Review the Skillstore skill "listening-script" from https://skillstore.io/skills/godgod126-listening-script.md and its manifest at https://skillstore.io/api/skills/godgod126-listening-script/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

你的智慧代理仍應顯示其計畫，並依安全政策請求必要確認。

Agent 可讀資源

當 AI Agent、爬蟲或腳本需要乾淨脈絡、而不是讀取完整頁面時，請使用這些連結。

Markdown 詳情 GET /skills/godgod126-listening-script.md 已簽名 manifest GET /api/skills/godgod126-listening-script/manifest 已簽名 lockfile GET /api/skills/godgod126-listening-script/lockfile

測試它

正在使用「listening-script」。 Adapt this fictional test into narration: twelve devices tested once per version; median waits were ten seconds for A and eight for B. Three devices favored A; battery use and user preferences were not measured.

預期結果:

Title: A Preliminary Comparison of Waiting Times
Narration: In a preliminary test, twelve devices ran the same passage once with each version.
Version A had a median wait of ten seconds, compared with eight seconds for version B.
That is a two-second reduction, or twenty percent relative to A.
Three devices were faster with A, so B was not faster on every device.
Battery use and user preferences were not measured.
Source notes: This example is fictional. It does not establish battery savings or user preference.

正在使用「listening-script」。 Adapt this supplied sentence for listening: The table above estimates an increase from 10 to 12 requests per minute in this sample.

預期結果:

Title: Estimated Request Rate
Narration: In this sample, the estimated request rate increased from ten to twelve requests per minute.
That is an increase of two requests per minute, or twenty percent relative to the starting rate.
Source notes: The estimate applies only to the stated sample. The supplied text gives no sample size or measurement date.

安全審計

安全

v1 • 2026/10/3 開啟版本化報告

Static analysis supplied no findings, so no finding verdicts are required. The two reviewed files describe text adaptation and explicitly prohibit credential requests and additional service uploads. No evidence found of prompt injection, data exfiltration, or unauthorized actions.

已掃描檔案

分析行數

複核項目

已忽略的誤報

最近一次完成的靜態與語意稽核未偵測到已確認的安全發現。這並不能證明該 Skill 沒有副作用。

審計單位: codex

分享與引用此報告

分享版本化評估報告、中性徽章、嵌入卡片與引用。Skillstore 只報告證據，不判斷該 Skill 是否安全。

開啟版本化報告

複製報告連結

https://skillstore.io/skills/godgod126-listening-script/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

[![Skillstore security assessment](https://skillstore.io/badges/skills/godgod126-listening-script/security.svg)](https://skillstore.io/skills/godgod126-listening-script?utm_source=security_passport_badge)

HTML 徽章

<a href="https://skillstore.io/skills/godgod126-listening-script?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/godgod126-listening-script/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

嵌入卡片

<iframe src="https://skillstore.io/embed/skills/godgod126-listening-script.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

學術引用 (APA · BibTeX · CFF)

APA 引用

godgod126. (2026). listening-script security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/godgod126-listening-script/audits/1

BibTeX 引用

@techreport{godgod126-godgod126-listening-script-2026,
  author = {godgod126},
  title = {listening-script security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/godgod126-listening-script/audits/1},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "listening-script security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "godgod126"
date-released: "2026-10-03"
url: "https://skillstore.io/skills/godgod126-listening-script/audits/1"
identifiers:
  - type: other
    value: "skillstore:godgod126-listening-script:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore 評分

評分依據證據置信度: 中

架構

可維護性

內容

社群

規範符合度

你可以打造什麼

Prepare article narration

Adapt a source-backed article for listening while retaining spoken attribution and moving detailed references into separate notes.

Explain research comparisons aloud

Convert report tables into spoken comparisons without removing sample limits, negative results, or uncertainty.

Create accessible study narration

Turn book notes or technical reading into consistent narration with clarified acronyms and verbal explanations of formulas.

試試這些提示

Adapt a short article

Adapt the supplied article into an English listening script. Preserve its claims and detail. Return a title, narration, and separate source notes. Source: [article]

Narrate a comparison table

Turn this paragraph and table into spoken narration for [audience]. Preserve units, baselines, sample sizes, exceptions, and uncertainty. Record omissions separately. Source: [text and table]

Adapt technical material

Adapt [technical source] into [language] narration for [audience]. Explain formulas verbally and retain exact forms in notes. Preserve attribution and qualifiers. Target length: [word count].

Maintain a long report across sections

Adapt [report] section by section using [glossary] and [voice]. Preserve detail and evidence limits. State coverage and omissions. Check numbers, causality, attribution, and qualifiers before delivery.

最佳實務

Supply source text and references, then specify audience, language, length, voice, and permitted editorial changes.
Keep narration separate from source notes, exact formulas, raw URLs, and unresolved editorial questions.
Review numbers, attribution, qualifiers, and omissions against the source before using the script.

避免

Requesting a listening adaptation while silently expecting major summarization or unsupported additions.
Turning a preliminary benchmark into a universal claim, product recommendation, or verified user outcome.
Treating a text script as synthesized audio or its estimated duration as a measured recording.

常見問題

Which AI tools support this skill?

The skill declares support for Claude, Codex, and Claude Code.

Does it generate audio?

No. It produces narration text that you can use separately with a text-to-speech tool or a human narrator.

Does it summarize the source automatically?

No. Making text listenable does not authorize summarization. Specify your desired length and any permitted compression.

How does it handle tables and formulas?

It converts tables into spoken relationships and explains formulas verbally. Exact formulas and omitted details can remain in separate notes.

Does it independently fact-check AI drafts?

No. It preserves source boundaries and flags unresolved claims. New research uses only available, authorized tools and sources.

How does it handle long documents?

It maintains continuity and glossary consistency across sections. If processing is partial, it states coverage rather than implying the entire document was adapted.