The Agent Skills Directory

[SAFE]: The skill implements a dedicated PII detection and redaction utility (src/lib/pii-detector.js). It automatically identifies and masks sensitive information such as emails and phone numbers in sample data before it is sent to external AI services.\n- [COMMAND_EXECUTION]: The tool utilizes the Treasure Data CLI (tdx) via the spawn command to interact with segments and schemas. Arguments are properly quoted to prevent injection, and the operations are transparently managed within the skill's source code.\n- [EXTERNAL_DOWNLOADS]: The skill connects to the Anthropic API to generate descriptions and the Treasure Data REST API (US and EU regions) to update schema comments. These connections target official vendor endpoints related to the tool's core functionality.\n- [SAFE]: Robust validation mechanisms (src/validation/csv-validator.js) are used to protect immutable schema properties, ensuring that column types and source identifiers cannot be modified during the review process.\n- [SAFE]: The inclusion of dry-run capabilities and a snapshot-based rollback system (src/lib/snapshot-manager.js) provides users with a safe environment to test and revert metadata changes.

data-dictionary