The Agent Skills Directory

[SAFE]: The skill performs terminal recording redaction using a bundled Python script (scripts/redact_cast.py) that relies solely on standard libraries and executes entirely on the local system without external dependencies or network access.\n- [SAFE]: The skill instructions specifically direct the AI agent to avoid reading raw recording files into its context, providing a strong defense against unintentional data exposure to telemetry or session logs.\n- [SAFE]: The Python script handles files securely by writing sanitized output to a new destination and verifying the result with a follow-up scan, ensuring original source files are never modified in place.\n- [SAFE]: No obfuscated code, hardcoded credentials, or suspicious execution patterns were identified within the skill's scripts or documentation.

cast-redact