event-driven
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a susceptibility to indirect prompt injection via untrusted data ingestion.\n
- Ingestion points: The skill uses the
read_urltool to fetch news and announcement text from external sources, which is then processed by the LLM.\n - Boundary markers: The prompt template directly interpolates the
{news_content}variable. It lacks specific delimiters or instructions for the LLM to ignore embedded commands or adversarial instructions within the news text.\n - Capability inventory: The agent generates numerical scores that are used by a Python script (
signal_engine.py) to calculate trading signals and manages results in a CSV format.\n - Sanitization: No input validation or sanitization is performed on the news content before it is passed to the LLM for sentiment analysis.
Audit Metadata