factor-crowding-monitor

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to execute shell commands for environment management and data retrieval using local vendor scripts.
  • Evidence: references/data-queries.md instructs the agent to run source ../../.venv/bin/activate and execute python ../findata-toolkit-cn/scripts/views_runner.py.
  • [EXTERNAL_DOWNLOADS]: The skill initiates the installation of Python packages via a requirements file associated with the vendor's toolkit.
  • Evidence: references/data-queries.md includes the command python -m pip install -r ../findata-toolkit-cn/requirements.txt.
  • [PROMPT_INJECTION]: The skill ingests untrusted market data, creating a surface for indirect prompt injection.
  • Ingestion points: External financial data from AKShare and East Money is fetched via views_runner.py (referenced in references/methodology.md).
  • Capability inventory: The skill has the capability to run shell commands and Python scripts.
  • Boundary markers: The skill employs an output template (references/output-template.md) but lacks explicit instructions to delimit or ignore instructions within external data.
  • Sanitization: There is no mention of data validation or sanitization for the ingested market content.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:23 AM