trader-alerts
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill performs network requests to fetch financial data from well-known services, specifically Yahoo Finance (query2.finance.yahoo.com) and Stooq (stooq.com). These operations are limited to retrieving ticker quotes and do not involve access to sensitive local files or credentials.
- [INDIRECT_PROMPT_INJECTION]: The skill represents an ingestion surface for external data as it parses JSON and CSV responses from external financial APIs. * Ingestion points: External data is ingested from the Yahoo Finance Chart API and the Stooq CSV endpoint. * Boundary markers: Explicit boundary markers or instructions to ignore embedded commands are absent in the alert messages generated from the fetched data. * Capability inventory: The skill writes to local files (trader/prices.json, trader/alerts.json) and outputs status messages to the agent, but does not perform shell command execution, code evaluation, or direct network exfiltration of local data. * Sanitization: The Python script (alerts.py) strictly parses the external data, casting price values to floating-point numbers and validating ticker symbols, which significantly mitigates the risk of processing malicious text payloads.
Audit Metadata