shareholder-structure-monitor

Warn

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute Python scripts from a relative path (../findata-toolkit-cn/scripts/views_runner.py) that is outside the skill's directory. This allows for the execution of code not contained within the audited package.
  • [EXTERNAL_DOWNLOADS]: The skill performs environment setup by running pip install on a requirements file located in a parent directory (../findata-toolkit-cn/requirements.txt), which involves downloading and installing unverified third-party dependencies.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its handling of external data.
  • Ingestion points: Aggregates data from multiple third-party financial platforms (EastMoney, Sina, THS, SSE, SZSE, BSE, CNINFO) and processes user-provided tables or screenshots (SKILL.md, references/data-queries.md).
  • Boundary markers: Absent. The logic does not specify delimiters or instructions to ignore embedded commands in the retrieved data.
  • Capability inventory: The skill has the ability to execute shell commands and Python scripts as part of its data retrieval workflow.
  • Sanitization: No validation or sanitization of the external data is performed before it is processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 10, 2026, 03:24 AM