skills/vm0-ai/vm0-skills/mercury/Gen Agent Trust Hub

mercury

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses curl to interact with the Mercury API for banking tasks like listing accounts and creating transfers.
  • [DATA_EXFILTRATION]: The skill accesses sensitive financial information, including account balances, transaction histories, and statements, directly from the official Mercury API domain (api.mercury.com).
  • [PROMPT_INJECTION]:
  • Ingestion points: Financial data returned from the API (such as transaction notes and invoice descriptions) is processed into the agent's context.
  • Boundary markers: None provided in the instructions.
  • Capability inventory: Shell command execution via curl.
  • Sanitization: None specified for external data received from the API.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 04:44 PM