handoff
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted conversation data to generate a summary and suggest future actions. Without explicit boundary markers, malicious instructions embedded in the conversation could influence the summary content or the "suggested skills" section provided to the next agent.
- Ingestion points: The skill ingests the "current conversation" as specified in SKILL.md.
- Boundary markers: There are no explicit delimiters (such as XML tags or unique markers) defined to isolate the conversation data from the skill's instructions.
- Capability inventory: The skill performs file system write operations to save the handoff document.
- Sanitization: The skill contains explicit instructions to redact sensitive information like API keys, passwords, and personally identifiable information (PII).
- [DATA_EXFILTRATION]: The skill instructs the agent to save the handoff document to the operating system's temporary directory rather than the current workspace. While this is likely intended for session management, it moves potentially sensitive summary data to a location outside the controlled project environment.
Audit Metadata