lov-better-statusline
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes natural language prompts to modify target application configuration files (status lines), creating an inherent attack surface for indirect injection.
- Ingestion points: User instructions and project facts provided via the natural language interface as described in
SKILL.md. - Boundary markers: The skill explicitly defines execution boundaries in
SKILL.md, requiring identification of target applications and state verification before modification. - Capability inventory: The skill utilizes host-provided CLI tools and file-writing capabilities to update configurations; it also uses
scripts/profile_store.pyfor local data persistence. - Sanitization: The
scripts/profile_store.pyscript implements a blacklist (SENSITIVE_PARTS) to filter out fields containing keywords like 'token', 'secret', or 'password', preventing their storage in the user profile records. - [COMMAND_EXECUTION]: The skill uses local scripts and host CLI tools to perform its primary functions, including backup, restoration, and configuration of status lines.
- Evidence:
scripts/profile_store.pyis used to manage local JSON profile data. The script utilizes atomic write operations viatempfile.NamedTemporaryFileandos.replaceto ensure data integrity and enforces restricted file permissions (0o600) to protect user data.
Audit Metadata