codex-context-details
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run a bundled Python script (
scripts/context_details.py) and standard Codex CLI tools (e.g.,codex mcp list,codex plugin list). These commands are used to gather diagnostic information about the local environment and the application's current state. - [DATA_EXFILTRATION]: The skill reads from local application telemetry files, including
~/.codex/logs_2.sqliteand JSONL session files. These files contain sensitive interaction history; however, the script's logic is specifically restricted to measuring byte sizes and counting tokens for reporting purposes. No network capabilities or external transmission patterns were identified in the script or instructions.
Audit Metadata