handoff
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes the conversation history, which is untrusted data that could contain malicious instructions designed to influence the agent's summary or subsequent actions. * Ingestion points: Conversation history processed in SKILL.md. * Boundary markers: Absent; there are no specific delimiters to differentiate between the agent's instructions and the untrusted conversation content. * Capability inventory: The agent is instructed to write to the file system (temporary directory). * Sanitization: The skill includes a proactive instruction to redact sensitive data such as API keys, passwords, and personally identifiable information.
- [DATA_EXFILTRATION]: The skill instructs the agent to save the resulting handoff document to the user's operating system temporary directory instead of the current project workspace. This is a functional requirement for the skill's purpose of generating a transient handoff file.
Audit Metadata