cgartlab-quill@0.13.0
VERIFIED· install onlyPublisher claims: "结构化长文写作框架：把模糊的写作意图整理成有核心论点、证据、受众、风格的长文，不绑定特定 harness。当用户要写博客、通讯、论文、专栏或任何非虚构长文，想把零散想法整理成有论点有证据的文章，或已有草稿却理不清主线时使用——只要意图是把想法变成有论点的长文就应激活本 skill，即使用户没明说「写文章」。不用于小说/诗歌/剧本等虚构创作或邮件/评论/即时消息等短文本。" Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Nothing executable in the bundle ran; the trace covers the install step only.- Tested
- every file read, file written and host contacted by what ran
- Not tested
- use of the skill by an agent
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (1 scenario; nothing executable in the bundle ran; the trace covers the install step only) on 2026-11-14 and scored it 87/100 (Verified). In what ran, nothing we could detect went wrong: no undeclared network egress, no credential reads, no persistent instruction writes. What it does with a real credential was not observed; the static findings and the publisher carry the rest of the score. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 0 s · 1 scenarios · 0 deviationsStatic findings
<summary>技术细节：为什么是重发而不是重试</summary> - 0.11.1 提交后 `publicationStatus: pending`；随后的 ClawScan 报 `Worker error: ClawScan scanner skillspector status was failed`，`wrFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary>技术细节：设计取舍与验证</summary> **为什么作者画像放在写作根目录、而不是加进文章七文件** 作者画像是**跨文章**知识（他一路怎么判断事情），而文章文件夹是**单篇**的。放进单篇会导致每篇都要重学、且画像随文章散落。放在根目录后，文章七文件约定完全不动，`_author-profilFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary>技术细节：ClawHub 对 0.11.0 的 8 条 findings 与本版处理</summary> 判定：`suspicious`（ClawScan / gpt-6.1-sol，置信度 high；SkillSpector 69 分 / DO_NOT_INSTALL；静态扫描另报 1 条 critFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary>技术细节：改了什么，怎么验证的</summary> **为什么这次能安全下沉禁令**：0.4.0 加过「时序」注记，但只覆盖风格检查；材料/来源闸门仍在写之前硬拦，所以「怕下手」。这次先把**两段式**立成原则（禁令明确属于校订期），再下沉禁令——**顺序反了会出问题**：先下沉而没有两段式，起草期就Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary>技术细节：与 0.5.0 的关系</summary> 0.5.0 建立了「不许编造经历与事件」的硬规则，来源是 **A 用户亲历 / B 公开来源 / C 无来源** 三分。0.5.1 **不放宽这条硬规则**（编造仍然禁止），只是**重新切分来源类别并移交取舍权**： - 旧的 A「用户亲历」只认「Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.