扫描报告
45 /100
session-reflect
Analyze Claude Code session history to generate self-observation journals
Skill declares local-only processing but commands use undocumented shell subprocess to cat config and sync sessions; also accesses undocumented ~/.claude memory path.
谨慎使用
Document all shell subprocess invocations in SKILL.md; clarify what data is read from ~/.claude memory path and why it's necessary.
安全发现 3 项
| 严重性 | 安全发现 | 位置 |
|---|---|---|
| 中危 | Undeclared shell execution in command files | commands/reflect/default.md, commands/reflect/drift.md, commands/reflect/emerge.md:3 |
| 中危 | Undeclared memory path access | commands/reflect/default.md:42 |
| 低危 | Hardcoded username in path | commands/reflect/default.md:42 |
| 资源类型 | 声明权限 | 推断权限 | 状态 | 证据 |
|---|---|---|---|---|
| 文件系统 | READ | READ | ✓ 一致 | commands/reflect/default.md:42 reads ~/.claude memory path |
| 命令执行 | NONE | WRITE | ✗ 越权 | Commands execute python3 and cat via shell |
| 环境变量 | NONE | NONE | — | No direct env access |
| 网络访问 | NONE | NONE | — | No network requests found |
2 项发现
中危 外部 URL 外部 URL
https://www.youtube.com/watch?v=6MBq1paspVU SKILL.md:115 中危 外部 URL 外部 URL
https://clawskills.sh/skills/riley-coyote-continuity SKILL.md:117 目录结构
6 文件 · 31.4 KB · 1069 行 Python 1f · 542L
Markdown 5f · 527L
├─
▾
commands
│ └─
▾
reflect
│ ├─
default.md
Markdown
│ ├─
drift.md
Markdown
│ └─
emerge.md
Markdown
├─
extract_sessions.py
Python
├─
reflect.md
Markdown
└─
SKILL.md
Markdown
安全亮点
✓ Sensitive data patterns (API keys, tokens, passwords) are properly redacted in session content
✓ All processing writes locally to user-specified Obsidian Vault
✓ No network exfiltration detected
✓ No credential harvesting or data theft indicators
✓ System prompts and boilerplate content are filtered
✓ Code blocks are replaced with placeholders