skills-panorama

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions direct the AI agent to execute its bundled shell script (bin/skill-panorama.sh), which orchestrates local system inspection via a Node.js CLI. This execution is scoped to local diagnostic operations provided within the skill package.
  • [DATA_EXPOSURE]: The tool accesses local skill directories and configuration files to generate its reports. It mitigates privacy risks by implementing an automated redaction system in lib/panorama-render.mjs that identifies and strips HOME paths, SSH endpoints, and potential credentials from reports intended for sharing.
  • [SAFE]: The skill architecture is explicitly read-only. It enforces strict boundaries against file system mutations and secures its local output files (latest.json, latest.md) with restricted Unix permissions (0600) to prevent unauthorized local access.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 05:47 AM
Security Audit — agent-trust-hub — skills-panorama