the-call-coach
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill handles untrusted transcripts (ingestion point: Mode B), creating an indirect injection surface. It mitigates this with clear boundary markers instructing the agent to treat input as data only and to quote rather than obey instructions. Capability inventory is restricted to text analysis and coaching report generation.- [CREDENTIALS_UNSAFE]: The skill proactively addresses the risk of credential exposure by instructing the agent to detect and redact secrets found in input transcripts or exports without reproducing them.
Audit Metadata