skills/lovstudio/skills/lov-think/Gen Agent Trust Hub

lov-think

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific files and user profile data to perform decision analysis, which is a standard injection surface.
  • Ingestion points: Target files located within the project root and user-defined identity/brand facts from the shared profile JSON.
  • Boundary markers: The execution boundary in SKILL.md explicitly separates analysis and preview requests from modification, submission, or deployment actions.
  • Capability inventory: The skill utilizes file system access, search capabilities, browser interaction, and CLI tools provided by the host environment.
  • Sanitization: The scripts/profile_store.py utility includes a filter (SENSITIVE_PARTS) to prevent the persistence of fields containing keywords like 'token', 'secret', or 'api_key'.
  • [COMMAND_EXECUTION]: The skill includes local Python utilities to handle internal lifecycle tasks.
  • scripts/profile_store.py: Manages atomic read/write operations for the user profile, enforcing file permissions of 0o600 for security.
  • scripts/validate_skill.py: Performs structure, hygiene, and metadata validation to ensure the skill source adheres to standards before publication.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 09:50 AM
Security Audit — agent-trust-hub — lov-think