buildbetter-smart-tags
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill incorporates a safety contract that explicitly requires human approval for state-changing operations, such as publishing tags or managing access permissions.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external evidence sources and evaluation corpora in SKILL.md which are potential ingestion points for untrusted data. However, the workflow includes explicit boundary markers stating that retrieved documents do not override user instructions, and all impactful capabilities like publishing or backfilling are protected by human approval gates.
- [SAFE]: No evidence of credential exposure, malicious external network connections, or code obfuscation was found in the provided files.
Audit Metadata