skills/stolinski/s-stack/handoff/Gen Agent Trust Hub

handoff

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use the mktemp command to generate a file path for storing handoff data on the local system.
  • [DATA_EXFILTRATION]: Conversation summaries are written to the local file system in temporary directories, which may result in data exposure of sensitive conversation details to other users or processes on the host.
  • [PROMPT_INJECTION]: The skill processes untrusted conversation history to create a summary for a future agent. This constitutes an indirect prompt injection surface where malicious content within the history could manipulate the behavior of the agent reading the handoff file.
  • Ingestion points: Conversation history (SKILL.md)
  • Boundary markers: Absent
  • Capability inventory: File write and path generation commands (SKILL.md)
  • Sanitization: Absent
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:55 PM
Security Audit — agent-trust-hub — handoff