setup
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from multiple tool outputs which presents an attack surface for indirect prompt injection.
- Ingestion points: Untrusted data is ingested into the agent context via the output of list_accounts, bridge_status, list_platforms, and get_help tools in SKILL.md.
- Boundary markers: The instructions do not define boundary markers or provide warnings to disregard instructions potentially embedded in the tool outputs.
- Capability inventory: The skill instructs the agent to display system commands (such as the wake command or mcp removal commands) to the user based on tool-provided information.
- Sanitization: There is no evidence of sanitization or validation of tool-returned data before it is presented to the user or used for procedural logic.
Audit Metadata