session-handoff
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill aggregates conversation history, file paths, and decisions into a 'paste-ready prompt' for a subsequent AI session.
- Ingestion points: The skill processes the entire current conversation context and inferred project state from the SKILL.md instructions.
- Boundary markers: While the output is formatted as markdown, there are no explicit instructions for the subsequent agent to treat the summarized content as potentially untrusted data or to ignore embedded instructions within the summary.
- Capability inventory: The generated prompt is intended to guide the behavior of a future agent session which may possess capabilities like file system access, network operations, or command execution.
- Sanitization: The skill does not explicitly sanitize the conversation history for potentially malicious prompt injection strings before including them in the handoff summary block.
Audit Metadata