handoff
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill creates a handoff document by summarizing the current conversation and user arguments. This process presents a risk of indirect prompt injection, as malicious content within the conversation or arguments could be transferred to the summary and influence the behavior of the next agent picking up the document.
- Ingestion points: 'current conversation' and user-supplied arguments (SKILL.md).
- Boundary markers: No delimiters or specific formatting are requested to isolate summarized content from agent instructions.
- Capability inventory: The skill involves writing the resulting document to the OS temporary directory.
- Sanitization: The instructions explicitly require the redaction of sensitive data like API keys and PII, which serves as a mitigation for data exposure but does not prevent instructional injection.
Audit Metadata