📚 Accusation — 指控
v1.0.0Audit Generator 在任何高风险谈判、艰难对话、薪资讨论、销售演示或……之前，生成一份预判异议审计与情绪标签库。
0· 16·0 当前·0 累计
by 安全扫描
OpenClaw
安全
high confidenceThis is an instruction-only negotiation assistant that coherently generates preemptive objection labels and scripts from user-provided conversation context and requires no installs, credentials, or unusual system access.
评估建议
This skill appears coherent and does not request credentials or installs, but it reads whatever conversation/context files you give it. Before installing or using it: (1) only provide the minimum context needed — redact bank account numbers, legal secrets, or unrelated personal data from situation-brief.md and counterpart-profile.md; (2) test the skill with a fictional or redacted scenario first to check tone and phrasing; (3) review all generated labels and scripts before using them in live con...详细分析 ▾
✓ 用途与能力
The skill's name and description match the SKILL.md: it asks for a situation brief and counterpart profile and produces an accusation audit and delivery script. It does not request unrelated binaries, services, or credentials.
ℹ 指令范围
Instructions direct the agent to read provided documents (situation-brief.md, counterpart-profile.md) and conversation history when available — this is appropriate for the stated purpose. Note: the skill requires access to user documents (Read/Write tools); ensure those documents do not contain secrets or unrelated private data. The guidance to surface 'embarrassing or extreme' accusations is consistent with the negotiation technique but could produce sensitive or strongly worded output, so review outputs before use.
✓ 安装机制
No install spec or code files are present (instruction-only). Nothing will be downloaded or written at install time by the skill itself.
✓ 凭证需求
The skill declares no environment variables, credentials, or config paths. Its required permissions (Read/Write documents) are proportional to the job of reading situation/context files and producing an audit.
✓ 持久化与权限
always is false and there are no claims to modify other skills or system-wide settings. The skill is user-invocable and can be invoked autonomously by the agent (platform default) — this is normal and not excessive for this kind of helper.
安全有层次，运行前请审查代码。
运行时依赖
无特殊依赖
版本
latestv1.0.02026/4/20
Initial release of Accusation Audit Generator. - Generate a preemptive objection audit and emotion-label bank before high-stakes conversations or negotiations. - Collects situation details and anticipated objections to produce an accusation audit artifact. - Converts anticipated negative emotions into neutral, third-person labels using the “It seems like…” formula. - Sequences emotion labels from strongest to lightest, with delivery script and pause instructions. - Designed for salespeople, founders, managers, and anyone expecting resistance or negative emotions in important conversations.
● 无害
安装命令
点击复制官方npx clawhub@latest install bookforge-accusation-audit-generator
镜像加速npx clawhub@latest install bookforge-accusation-audit-generator --registry https://cn.longxiaskill.com