dark-pool-routing
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided data via
scripts/dark_pool_routing_diagnostics.py, which reads aninput.csvfile usingpandas. This creates a data ingestion surface; however, the script performs strict numeric and datetime type enforcement on the input data, which serves as an inherent control against instruction interpretation. There are no mechanisms in the skill that would cause the agent to execute or evaluate the content of the CSV as instructions. - [COMMAND_EXECUTION]: The skill instructions provide a command to run a Python-based diagnostic tool (
scripts/dark_pool_routing_diagnostics.py). Analysis of the script confirms it is limited to performing statistical calculations (mean, standard deviation, percentiles) on the input CSV and writing a summary JSON file, matching the described purpose of the skill.
Audit Metadata