Scan Report
5 /100
SwarmClaw Platform
AI agent runtime and multi-agent orchestration platform documentation
SwarmClaw platform documentation describing a legitimate AI agent runtime with clearly declared capabilities and no malicious behavior.
Safe to install
No action required. This is a documentation file for a legitimate AI agent platform.
| Resource | Declared | Inferred | Status | Evidence |
|---|---|---|---|---|
| Filesystem | READ/WRITE | READ/WRITE | ✓ Aligned | files tool described in 'The 6 Primitive Tools' table |
| Shell | WRITE | WRITE | ✓ Aligned | execute tool described: 'Run bash scripts (sandboxed or host)' |
| Network | READ | READ | ✓ Aligned | execute tool with curl, browser tool for dynamic web scraping |
| Environment | READ | READ | ✓ Aligned | Credentials section: 'Injected as environment variables into execute tool runs' |
| Browser | READ/WRITE | READ | ✓ Aligned | browser tool described in 'The 6 Primitive Tools' table |
File Tree
1 files · 5.8 KB · 132 lines Markdown 1f · 132L
└─
SKILL.md
Markdown
Security Positives
✓ Pure documentation file with no executable code
✓ All tool capabilities are explicitly declared in clear documentation
✓ Credential handling is well-documented with automatic redaction
✓ No hidden functionality or suspicious patterns detected
✓ Best practices section promotes transparency and proper error handling
✓ File access boundaries and workspace conventions are clearly defined