handoff
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill summarizes current conversation history for consumption by a future agent session.
- Ingestion points: Current conversation history and user arguments (SKILL.md).
- Boundary markers: None defined in the instructions to separate the generated summary from potentially malicious content within the history.
- Capability inventory: No executable scripts provided; the skill relies on the agent's native capabilities to save the document to the OS temporary directory.
- Sanitization: The skill explicitly instructs to redact sensitive data (API keys, passwords, PII), which is a security best practice, but lacks instructions to sanitize or escape instructions that might be embedded in the conversation data.
- [NO_CODE]: The skill consists entirely of natural language instructions and configuration metadata. It contains no scripts, binaries, or automated command executions.
Audit Metadata