sz-ink@0.1.0
VERIFIEDPublisher claims: "研墨(sz-ink)——中文文本去AI味改写器:基于汉语言文学研究(语体学/词汇学/修辞学/语篇学)提供鉴别力, 不靠禁词表和同义词硬换。整体体检(九项统计指标+语体条件阈值)→ 定点干预(每轮≤3处, 按"症状→判据→动作→对照例"处方卡执行)→ 语义保真审核(锚点+论点清单,漂移即回滚)→ 回测交付。可区分黑话型/文案腔型/洗稿型三种AI味;任意文体均可输入(探店/博客/周报/信件乃至 单段片段,三语体就近归属,片段自动降级),支持按用户自带样本提取风格档案("像谁")。 Use when:用户给出现成的 AI 生成或 AI 腔明显的中文文本,要求"去AI味/改自然/像人写的/降AI感"; 或用户提供自己写的段落做风格参照后要求按其风格改写。Skip for:从零代写新文章(那是写作类 skill 的活);英文本改写;诗歌、法律条文、合同等不允许改动表达的文本;用户只想检测AI率不改稿 (本 skill 改写优先)。" Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (1 scenario) on 2026-10-29 and scored it 93/100 (Verified). It did what it claims and nothing else we could detect: no undeclared network egress, no credential reads, no persistent instruction writes. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 0 s · 1 scenarios · 0 deviationsStatic findings
No findings across 2,467 lines in 31 files.