handoff
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the conversation history to generate a summary. This presents a surface for indirect prompt injection where malicious content in the history could influence the summary or the "suggested skills" section, potentially leading to a subsequent agent executing unintended tools.
- Ingestion points: Conversation history and user arguments are processed in
SKILL.md. - Boundary markers: Absent; no explicit delimiters or boundary markers are defined for the summarized content.
- Capability inventory: The skill performs file-write operations to the system's temporary directory (
$TMPDIR,/tmp,%TEMP%). - Sanitization: Present; includes explicit instructions to redact sensitive information (API keys, passwords, PII) before writing.
Audit Metadata