tgw-onboarding
Pass
Audited by Gen Agent Trust Hub on Jul 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill incorporates a robust safety gate that detects when a user might be transitioning from practice to real business tasks (e.g., providing real brand names or file paths) and requires explicit confirmation before ending the tutorial mode to prevent accidental data persistence during learning.
- [PROMPT_INJECTION]: The skill features a discovery phase that reads content from local files to identify available tools, which represents an indirect prompt injection surface.
- Ingestion points: Reads content from
skills/tgw-*/SKILL.mdduring the startup scan in the SKILL.md file. - Boundary markers: No specific delimiters or "ignore instructions" wrappers are applied to the content read from these external skill files during the discovery process.
- Capability inventory: Across all scripts, the skill's capabilities are limited to reading local skill definitions and writing progress data to the specific path
tgw-data/learning/current-user.md. - Sanitization: There is no evidence of sanitization or filtering for the content ingested during the skill discovery scan.
Audit Metadata