lov-meta-command

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes historical command data and migration source files, which constitutes a surface for processing untrusted content.
  • Ingestion points: Historical slash command data, input/output contracts, and existing skill files processed during the conversion workflow.
  • Boundary markers: The SKILL.md execution boundary explicitly separates read-only analysis and preview phases from modification and deployment tasks.
  • Capability inventory: The skill utilizes bundled Python scripts (profile_store.py, validate_skill.py) for file management and metadata validation, and relies on the agent host's CLI and search tools.
  • Sanitization: The profile management utility implements a blocklist for sensitive field names such as 'token', 'secret', and 'api_key' to prevent accidental credential persistence in user-owned profiles.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 09:49 AM
Security Audit — agent-trust-hub — lov-meta-command