handoff
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes and summarizes the current conversation history. Because this history can contain data from untrusted sources (such as external web pages or files), there is a possibility that malicious instructions embedded in that data could be included in the summary.\n
- Ingestion points: The agent context containing the current conversation history (SKILL.md).\n
- Boundary markers: No specific boundary markers or instructions to ignore embedded commands are specified for the summary content.\n
- Capability inventory: The agent writes the summarized document to a file.\n
- Sanitization: The skill provides clear instructions to redact API keys, passwords, and personally identifiable information (PII).\n- [SAFE]: The skill instructs the agent to save the handoff document to the operating system's temporary directory. While this moves data outside the current workspace, it is a standard approach for persisting data across different sessions or projects.\n- [SAFE]: Explicit instructions are provided to redact sensitive information before saving the document, which reduces the risk of credential leakage in shared environments.
Audit Metadata