ib-account

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions provide specific shell commands to execute the scripts/account.py file using uv run python or python. This is the intended mechanism for the agent to retrieve account data.
  • [INDIRECT_PROMPT_INJECTION]: The skill retrieves account summaries from an external source (Interactive Brokers API) which are then processed by the agent.
  • Ingestion points: Account data returned as JSON by the scripts/account.py script (e.g., net liquidation, cash, buying power).
  • Boundary markers: The output is explicitly structured as JSON, which helps the agent distinguish between data and instructions.
  • Capability inventory: The skill allows local script execution and reading of environment variables for configuration.
  • Sanitization: The script outputs structured JSON with specific fields (accounts, connected), limiting the surface area for unexpected text processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 05:55 PM
Security Audit — agent-trust-hub — ib-account