handoff
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the conversation history to generate a handoff document, which could theoretically propagate malicious instructions if they were present in the source context.
- Ingestion points: All current conversation context used to populate the handoff sections.
- Boundary markers: The skill enforces a specific Markdown structure with defined headers to delimit context types.
- Capability inventory: The skill is restricted by
disable-model-invocation: true, which prevents the agent from calling tools or executing commands during the summary generation. - Sanitization: Instructions explicitly require the redaction of secrets, API keys, tokens, passwords, and PII from the output.
Audit Metadata