handoff
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted conversation history to generate a handoff document for a downstream agent. This creates a multi-step chain where malicious instructions in the chat could be carried over into the next agent's context. * Ingestion points: The entire conversation history is processed to create the handoff document. * Boundary markers: The skill does not specify using delimiters or 'ignore embedded instructions' warnings for the generated content. * Capability inventory: The skill performs file-write operations to the system temporary directory. * Sanitization: The skill explicitly instructs the agent to redact API keys, passwords, and PII, which serves as a basic sensitive data filter.
- [DATA_EXFILTRATION]: The skill instructs the agent to save data to the OS temporary directory, which is outside the current workspace, potentially exposing session summaries if the temporary directory is not properly secured.
Audit Metadata