user-context

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • Local Script Execution: The skill utilizes two Python scripts (user_context_preflight.py and init_user_context.py) to manage a local state file. These scripts handle file path resolution and directory creation within the user's environment to maintain persistent context between sessions.
  • Data Storage Considerations: The skill is designed to write and read from $CODEX_HOME/state/plugins/product-design/. While this involves file system operations, the instructions explicitly caution the agent against saving secrets, credentials, or private tokens, which is a key safety measure for context-persistence skills.
  • User-Controlled Data Ingestion: The skill processes user-provided links and images. It employs a preflight script to parse this data into a structured format, which helps ensure the information is handled consistently.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 09:17 AM
Security Audit — agent-trust-hub — user-context