buildbetter-backfill-smart-tag

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely instructional and describes a workflow using internal platform tools (MCP tools). It does not contain any executable scripts, shell commands, or external network dependencies.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process data from Smart Tag groups. It mitigates risk by enforcing a 'preview' step and requiring explicit user approval for operations, especially those involving billing or large datasets. It uses structured API tools rather than arbitrary shell or file system operations.
  • [DATA_EXPOSURE]: No sensitive information or credentials are hardcoded. The skill follows best practices by using idempotency UUIDs and bounded record counts.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 09:33 AM