扫描报告
0 /100
agent-provenance
Track authorship, review status, and governance of agent instruction files. Adds provenance headers, commit conventions, TTL on agent-written goals, and periodic diff reports.
This is a documentation-only skill describing a governance framework for tracking AI agent instruction file authorship. There are no scripts, code files, or dependencies — only a 108-line SKILL.md describing policy and conventions.
可以安装
No action needed. This skill is purely advisory documentation and poses no security risk.
| 资源类型 | 声明权限 | 推断权限 | 状态 | 证据 |
|---|---|---|---|---|
| 文件系统 | NONE | NONE | — | No file operations in the skill |
| 网络访问 | NONE | NONE | — | No network calls in the skill |
| 命令执行 | NONE | NONE | — | No shell execution in the skill |
| 环境变量 | NONE | NONE | — | No environment access in the skill |
| 技能调用 | NONE | NONE | — | No skill invocation in the skill |
| 剪贴板 | NONE | NONE | — | No clipboard access in the skill |
| 浏览器 | NONE | NONE | — | No browser access in the skill |
| 数据库 | NONE | NONE | — | No database access in the skill |
目录结构
1 文件 · 4.7 KB · 108 行 Markdown 1f · 108L
└─
SKILL.md
Markdown
安全亮点
✓ Skill is documentation-only with zero executable code
✓ No dependencies or package files present
✓ No sensitive file access described or implemented
✓ No network, shell, or credential operations
✓ No obfuscation or suspicious patterns detected
✓ The described functionality (provenance headers, TTL on goals, diff reports) is benign governance tooling