handoff
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes the current conversation (untrusted input) to generate a summary for a future agent. Malicious instructions or data embedded in the conversation could be persisted into the handoff document, potentially influencing the behavior of the next agent that reads it.
- Ingestion points: The entire current conversation history is ingested as source material for the handoff document.
- Boundary markers: The skill does not define specific boundary markers or delimiters for the handoff content to prevent the next agent from interpreting summary text as instructions.
- Capability inventory: The skill is capable of writing files to the operating system's temporary directory (
/tmpor%TEMP%). - Sanitization: The skill contains explicit instructions to redact sensitive information such as API keys, passwords, and personally identifiable information (PII) before saving the document.
Audit Metadata