claude-settings-audit
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The agent uses
ls,find, andcatto identify project files and existing configurations. These are read-only operations for environmental discovery. - [EXTERNAL_DOWNLOADS]: Installation instructions reference
npxto download the skill from the author's GitHub repository. This is a recognized installation method for the toolset. - [PROMPT_INJECTION]: The skill ingests untrusted data by reading repository manifests which presents an indirect prompt injection surface.
- Ingestion points: Reads repository dependency and configuration files using
cat. - Boundary markers: None.
- Capability inventory: Read-only discovery commands and recommendations for CLI tool access including
gh api:*. - Sanitization: None.
Audit Metadata