search-performance-optimizer
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements strong security boundaries by explicitly instructing the agent to never request credentials, tokens, or raw customer data. It further limits risks by stating that search execution or configuration changes should not be performed without explicit user authorization.
- [SAFE]: The skill incorporates clear instructions to mitigate indirect prompt injection by treating all user-supplied evidence (such as SPL and logs) as untrusted text/evidence rather than instructions.
- [SAFE]: External downloads are restricted to verifying official Splunk documentation from a well-known, trusted domain (help.splunk.com), which aligns with the skill's stated purpose and the author's identity.
Audit Metadata