中文 EN

Scan Report

Scan New Files

⚠ Low Risk — Risk Score 15/100
Last scan:2 days ago Rescan
15 /100
clawscan-vigil
OpenClaw Skill Security Scanner - static and dynamic analysis
ClawScan is a legitimate security scanning tool for OpenClaw Skills with no malicious behavior detected. All capabilities align with documented functionality.
Skill Nameclawscan-vigil
Duration51.1s
Enginepi
✓
Safe to install
Approve for installation. This is a genuine security tool that performs static and dynamic analysis of AI skills.

Findings 2 items

Severity Finding Location
Low
Client-side license validation
License verification is performed locally without server verification. The 'verified' flag remains False even after activation.
'verified': False  # Would be True after server verification
→ Implement server-side license verification for production use
 core/license_manager.py:131
Info
RestrictedPython sandbox limitations
Dynamic analysis uses RestrictedPython which may not catch all runtime behaviors. Some malicious code patterns skip detection.
from RestrictedPython import compile_restricted, safe_globals
→ Document sandbox limitations in SKILL.md
 core/dynamic_tracer.py:1
ResourceDeclaredInferredStatusEvidence
Filesystem READ READ ✓ Aligned Reads Python files in target skill directories for analysis
Shell WRITE WRITE ✓ Aligned skill/skill_wrapper.py:17 uses subprocess.run() to invoke clawscan CLI
Network NONE READ ✓ Aligned requests library is a dependency but used only for legitimate external verificat…
8 findings
🔗
Medium External URL 外部 URL
https://clawhub.ai/signup
PUBLISH.md:45
🔗
Medium External URL 外部 URL
https://img.shields.io/badge/python-3.10+-blue.svg
README.md:3
🔗
Medium External URL 外部 URL
https://www.python.org/downloads/
README.md:3
🔗
Medium External URL 外部 URL
https://img.shields.io/badge/License-MIT-yellow.svg
README.md:4
🔗
Medium External URL 外部 URL
https://opensource.org/licenses/MIT
README.md:4
🔗
Medium External URL 外部 URL
https://clawscan.dev
SKILL.md:73
🔗
Medium External URL 外部 URL
https://docs.clawscan.dev
pyproject.toml:42
📧
Info Email 邮箱地址
[email protected]
SKILL.md:167

File Tree

16 files · 74.9 KB · 2421 lines
Python 11f · 1751L Markdown 4f · 616L TOML 1f · 54L
├─ ▾ 📁 core
│ ├─ 🐍 __init__.py Python 5L · 176 B
│ ├─ 🐍 advanced_analyzer.py Python 230L · 8.1 KB
│ ├─ 🐍 batch_scanner.py Python 156L · 5.3 KB
│ ├─ 🐍 dynamic_tracer.py Python 249L · 9.0 KB
│ ├─ 🐍 license_manager.py Python 159L · 5.0 KB
│ ├─ 🐍 models.py Python 59L · 1.6 KB
│ ├─ 🐍 risk_engine.py Python 110L · 3.8 KB
│ ├─ 🐍 scanner.py Python 101L · 3.4 KB
│ └─ 🐍 static_analyzer.py Python 262L · 9.3 KB
├─ ▾ 📁 skill
│ ├─ 🐍 skill_wrapper.py Python 74L · 1.9 KB
│ └─ 📝 SKILL.md Markdown 171L · 4.2 KB
├─ 🐍 cli.py Python 346L · 11.0 KB
├─ 📝 PUBLISH.md Markdown 135L · 2.5 KB
├─ 📄 pyproject.toml TOML 54L · 1.3 KB
├─ 📝 README.md Markdown 139L · 4.0 KB
└─ 📝 SKILL.md Markdown 171L · 4.2 KB

Dependencies 4 items

PackageVersionSourceKnown VulnsNotes
click >=8.0 pyproject.toml No CLI framework - standard dependency
requests >=2.28 pyproject.toml No HTTP library - expected for network security scanning
RestrictedPython >=7.0 pyproject.toml No Security sandbox - core to this tool's safety model
rich >=13.0 pyproject.toml No Terminal output - standard UI dependency

Security Positives

✓ Uses RestrictedPython sandbox for dynamic analysis
✓ AST-based static analysis detects dangerous patterns
✓ No credential harvesting or exfiltration code found
✓ No base64/eval patterns that could hide malicious behavior
✓ Subprocess usage is scoped to legitimate tool execution
✓ All declared capabilities match actual implementation
✓ No access to ~/.ssh, ~/.aws, or other sensitive credential paths
✓ Dependencies (click, rich, requests) are standard and legitimate