handoff
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill contains no malicious code or instructions. It facilitates session transitions by summarizing work state in a structured manner.
- [SAFE]: The skill incorporates specific rules to redact sensitive information like API keys and PII, and avoids saving to the project workspace to prevent accidental data exposure.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted conversation data but mitigates risks through explicit redaction and restricted output paths. (Ingestion: Conversation history; Boundary markers: Redaction instructions present; Capability: Local file write to temporary directory; Sanitization: Explicit redaction rules).
Audit Metadata