crossing-liquidity-access
Pass
Audited by Gen Agent Trust Hub on Apr 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions and documentation focus exclusively on financial trading workflows without attempting to override agent safety protocols or system behavior.
- [COMMAND_EXECUTION]: The skill utilizes a local Python script for deterministic data analysis. The script performs standard data manipulation using the pandas library and handles local file I/O for reading CSV inputs and writing JSON results. There is no evidence of arbitrary or dangerous command execution.
- [EXTERNAL_DOWNLOADS]: The Python diagnostic script depends on the well-known 'pandas' library for data processing. No remote scripts or unverified third-party packages are downloaded or executed.
- [DATA_EXFILTRATION]: No network operations or credential harvesting patterns were found. Data processing is restricted to the local environment.
- [PROMPT_INJECTION]: The skill ingests data via 'input.csv' in 'scripts/crossing_liquidity_access_diagnostics.py'. No boundary markers are present in the documentation for this ingestion point as the script is handled as a deterministic tool. Capability inventory is limited to local file reading and writing with no network or dynamic execution. Sanitization is performed via numeric and date coercion within the Python script logic.
Audit Metadata