moomoo-capital-anomaly
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script (scripts/handle_capital_anomaly.py) using shell commands with arguments derived from user input. The script is invoked using the standard python interpreter found in the environment.
- [PROMPT_INJECTION]: The skill processes untrusted user input for stock names and time ranges, creating an indirect prompt injection surface.
- Ingestion points: User-provided stock names, tickers, and company names in SKILL.md workflow.
- Boundary markers: Absent.
- Capability inventory: Subprocess execution of a local Python script.
- Sanitization: The skill instructions mandate a normalization step where the agent maps loose user mentions to standard market-prefixed symbols before script execution, which serves as an effective validation layer.
- [SAFE]: The Python script performs dynamic path modification to import the futu library from its parent directory. This is a standard method for resolving local dependencies in this skill's architecture.
- [SAFE]: Network activity is directed exclusively to the local loopback address (127.0.0.1:11111), which is the default endpoint for the Futu OpenD API gateway.
Audit Metadata