fleet

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands including git, find, and ls, as well as inline Python scripts to gather node data and file metadata.
  • [EXTERNAL_DOWNLOADS]: The skill performs network operations by using the 'maw' tool to send requests to remote addresses identified in the contacts list.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface. Ingestion points: contacts.json, settings.json, and remote responses via maw. Boundary markers: Absent. Capability inventory: File system access (find, ls), repository metadata (git), and network messaging (maw). Sanitization: None.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 09:19 PM
Security Audit — agent-trust-hub — fleet