handoff
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (cross-session injection).
- Ingestion points: The skill ingests the 'current conversation' (untrusted data) from the environment in SKILL.md.
- Boundary markers: Absent. There are no instructions to use delimiters or to warn the agent to ignore embedded instructions within the conversation history.
- Capability inventory: The skill triggers file-write operations to the OS temporary directory.
- Sanitization: The skill explicitly requires the redaction of sensitive credentials (API keys, passwords) and PII, which is a strong safety measure. However, it lacks sanitization or validation for the 'suggested skills' section, which allows malicious instructions from the conversation history to influence the behavior of the next agent session.
Audit Metadata