handoff
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from the current conversation and user arguments to generate a handoff document saved to the local file system. This creates a surface for indirect prompt injection where malicious content in the conversation could influence the summary or affect downstream agents reading the file.
- Ingestion points: Current conversation history and user-supplied arguments defined in
SKILL.md. - Boundary markers: None present to delimit or isolate untrusted session data from the summarization instructions.
- Capability inventory: File-write operations to the
.scratchpad/handoff/directory as specified inSKILL.md. - Sanitization: The skill includes proactive security instructions to redact sensitive information such as API keys, passwords, and PII from the output, which mitigates data exposure risks.
Audit Metadata