statusline-install
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill demonstrates a surface for indirect prompt injection by processing and displaying external data without explicit sanitization.
- Ingestion points: Metadata fields such as
.model.display_name,.cost.total_cost_usd, and.context_window.total_input_tokensdefined inSKILL.md. - Boundary markers: Absent; the skill interpolates data directly into the status line string.
- Capability inventory: The skill requires the ability to write to
~/.claude/settings.jsonto persist the configuration. - Sanitization: Not present; the instructions do not specify escaping or validation for the data sources before they are rendered in the terminal.
Audit Metadata