news-analytics-trading
Pass
Audited by Gen Agent Trust Hub on Apr 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script
scripts/news_analytics_trading_diagnostics.pyas part of its core workflow to process quantitative data. - [DATA_EXFILTRATION]: The diagnostics script reads from local CSV files and writes summary results to JSON files. While focused on performance metrics, this creates a handling surface for local file data.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via processed data files.
- Ingestion points: The script
scripts/news_analytics_trading_diagnostics.pyreads untrusted data from external CSV files usingpd.read_csv. - Boundary markers: No boundary markers or instructions are used to distinguish data from instructions within the processed CSV content.
- Capability inventory: The skill possesses file read and write capabilities through the
pandaslibrary and standard file I/O operations. - Sanitization: Input data, including column headers and values, are processed and returned in a summary JSON without sanitization or escaping, allowing potentially malicious strings to enter the agent's context.
Audit Metadata