lov-open-codex-session

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local Python script (profile_store.py) to manage user profile state. This script is called by the agent to read and persist configuration data. The script uses standard Python libraries and includes regex-based validation for identifiers and a blacklist for sensitive field names.
  • [DATA_EXPOSURE]: The skill implements a local storage mechanism for user profiles, typically located in ~/.lovstudio/skills/profile.json. This is used to store non-sensitive preferences like language and timezone. The implementation includes logic to prevent the persistence of fields containing strings such as 'token', 'secret', or 'password'.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes session identifiers and user-provided profile values. It mitigates potential injection by validating input formats (UUIDs) and restricting profile updates to specific namespaces and keys.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 12:59 AM
Security Audit — agent-trust-hub — lov-open-codex-session