可信 — 风险评分 0/100
上次扫描:1 天前 重新扫描
0 /100
ContextLedger
Evidence-first knowledge auditing skill that upgrades connected knowledge into an auditable conclusion card. Traces sources, marks staleness, surfaces conflicts, and separates evidence from inference.
ContextLedger is a purely documentation-based skill with no executable code, dependencies, or sensitive file access. It consists only of markdown documentation describing a knowledge auditing workflow.
技能名称ContextLedger
分析耗时29.7s
引擎pi
可以安装
This skill is safe to use. No security concerns identified.
资源类型声明权限推断权限状态证据
文件系统 NONE NONE No executable code; only markdown documentation files present
网络访问 NONE NONE No network requests or web fetches in any file
命令执行 NONE NONE No shell scripts or subprocess calls present
环境变量 NONE NONE No environment variable access in any file
技能调用 READ READ ✓ 一致 SKILL.md: Referenced by other skills like 'Knowledge Connector', 'DecisionDeck',…
剪贴板 NONE NONE No clipboard access documented or implemented
浏览器 NONE NONE No browser automation present
数据库 NONE NONE No database access in any file

目录结构

8 文件 · 21.3 KB · 690 行
Markdown 6f · 662L JSON 1f · 24L YAML 1f · 4L
├─ 📁 agents
│ └─ 📋 openai.yaml YAML 4L · 405 B
├─ 📁 references
│ ├─ 📝 audit-heuristics.md Markdown 113L · 3.3 KB
│ └─ 📝 conclusion-cards.md Markdown 74L · 1.3 KB
├─ 📝 CHANGELOG.md Markdown 15L · 841 B
├─ 📋 clawhub.json JSON 24L · 606 B
├─ 📝 README.md Markdown 86L · 2.9 KB
├─ 📝 RELEASE.md Markdown 106L · 3.0 KB
└─ 📝 SKILL.md Markdown 268L · 9.0 KB

安全亮点

✓ No executable code - skill consists entirely of documentation files
✓ No dependencies or package management files
✓ No sensitive file access (no .env, ~/.ssh, ~/.aws, etc.)
✓ No network requests or data exfiltration potential
✓ No shell execution, subprocess, or eval patterns
✓ SKILL.md clearly documents all intended behavior
✓ No obfuscation or base64-encoded content
✓ Clean file inventory: 6 markdown files, 1 JSON config, 1 YAML agent config