z-session-handoff
Pass
Audited by Gen Agent Trust Hub on Aug 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes current conversation history (untrusted data) to generate a handoff document that is explicitly designed to be read and acted upon by a future AI agent session.
- Ingestion points: The skill ingests the entirety of the current conversation history to synthesize goals, states, and decisions (SKILL.md).
- Boundary markers: The skill requires an explicit 'HISTORICAL REFERENCE' banner and mandates that past commands be quoted as facts rather than imperatives to reduce the risk of the next agent re-executing stale or malicious instructions.
- Capability inventory: The skill utilizes file-writing capabilities to save the handoff document to the operating system's temporary directory.
- Sanitization: The instructions explicitly require the redaction of API keys, passwords, tokens, and personal data before saving the document.
Audit Metadata