Skills/ClawHub/andrew-translate

andrew-translate@1.3.2

VERIFIED

Publisher claims: "高质量三步式翻译工作流（初稿 → 专家反思 → 精稿），基于 Andrew Ng 的 translation-agent 思路，产出接近母语者水准的译文。凡是用户要翻译成段文本（中英互译或任意语种；文章、技术文档、论文、书籍、邮件、字幕、产品文案、代码注释），或要求"高质量翻译""专业翻译""信达雅""润色/校对已有译文""三步翻译""反思式翻译""translation agent"时都应使用本 skill；即使用户只说"帮我翻译这段话""把这段英文翻成中文""translate this"，也应使用它而不是直接给一遍机翻。需要指定地区风格（台湾繁体、香港繁体、美式/英式英语、墨西哥西班牙语、巴西葡语等）时尤其适用。即使原文还没贴出、文件不在工作区，也先读本 skill 再处理。单个词义查询或术语解释不需要本 skill。" Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry ClawHubPublisher linkxzhou · joined 0 days agoInstalls 333Last detonated 2 h ago
$clawhub install andrew-translate
WatchReport an issue

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (1 scenario) on 2026-10-27 and scored it 82/100 (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 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
0.0sprocessscript backend runs setup only
0.0sprocessscript backend runs setup only

Static findings

CRITICAL
Prompt injection attemptevals/evals.json:54 · Prompt injection attempt — 'ignore previous instructions'Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST01
MEDIUM
Process kill commandevals/evals.json:60 · Process kill command — agent can terminate processesFix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
AST05