insight-brief-generator 安全扫描报告 — 可信 | ClawSafe

0 /100

insight-brief-generator

把报表和图表转成管理层可读的洞察摘要，区分发现、解释和建议动作

This skill is a straightforward business-report summarizer with no malicious behavior, credential access, network exfiltration, or obfuscation. Shell execution via python3 is declared, filesystem writes are limited to output files, and all operations are audit-friendly.

技能名称insight-brief-generator

分析耗时31.3s

引擎pi

✓

可以安装

Approve for use. No security concerns detected. The only filesystem WRITE is the explicitly named output file.

资源类型	声明权限	推断权限	状态	证据
文件系统	`READ`	`READ`	✓ 一致	scripts/run.py — only reads input files; output write is user-specified via --ou…
命令执行	`EXEC`	`EXEC`	✓ 一致	SKILL.md line 21: `python3 "{baseDir}/scripts/run.py" --input ...`
网络访问	`NONE`	`NONE`	—	No network calls in scripts/run.py — uses only Python standard library
环境变量	`NONE`	`NONE`	—	run.py uses os.environ only to access PATH for resolution; no credential harvest…
技能调用	`NONE`	`NONE`	—	No cross-skill invocation
剪贴板	`NONE`	`NONE`	—	No clipboard access
浏览器	`NONE`	`NONE`	—	No browser access
数据库	`NONE`	`NONE`	—	No database access

1 项发现

🔗

中危外部 URL 外部 URL

https://example.invalid/skills/insight-brief-generator

SKILL.md:6

目录结构

9 文件 · 20.6 KB · 561 行

Python 1f · 282L Markdown 7f · 243L JSON 1f · 36L

├─ ▾ 📁 examples

│ ├─ 📝 example-input.md Markdown 18L · 607 B

│ └─ 📝 example-output.md Markdown 24L · 613 B

├─ ▾ 📁 resources

│ ├─ 📋 spec.json JSON 36L · 900 B

│ └─ 📝 template.md Markdown 23L · 883 B

├─ ▾ 📁 scripts

│ └─ 🐍 run.py Python 282L · 11.1 KB

├─ ▾ 📁 tests

│ └─ 📝 smoke-test.md Markdown 34L · 885 B

├─ 📝 README.md Markdown 70L · 2.2 KB

├─ 📝 SELF_CHECK.md Markdown 19L · 1.1 KB

└─ 📝 SKILL.md Markdown 55L · 2.3 KB

依赖分析 1 项

包名	版本	来源	已知漏洞	备注
`python3`	`any (runtime)`	system	否	Declared in metadata.openclaw.requires.bins; no third-party packages

安全亮点

✓ No credential harvesting — PATTERNS dict contains 'secret_like' only as a scan regex, not for exfiltration

✓ No network requests — script uses only Python standard library, no urllib/requests/httpx

✓ No obfuscation — no base64, no eval(), no encoded strings

✓ No sensitive path access — does not touch ~/.ssh, ~/.aws, .env, or key directories

✓ No remote script execution — no curl|bash, no wget|sh, no download-and-run patterns

✓ Output file write is opt-in and user-specified via --output flag

✓ Shell execution is documented in SKILL.md

✓ Dependencies: only python3 + Python standard library (fully declared in metadata.openclaw.requires.bins)

✓ Includes --dry-run mode for read-only analysis without file writes

✓ All modes (structured_brief, directory_audit, csv_audit, pattern_audit, skill_audit) are for generation and auditing — no execution of found patterns