us-market-bubble-detector
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
- Ingestion points: The skill instructs the agent to use
web_searchto collect real-time financial metrics (e.g., Put/Call ratio, VIX, Margin Debt, IPO stats) as defined inPhase 1: Mandatory Quantitative Data Collectionin bothSKILL.mdandREADME.md. - Boundary markers: The instructions lack explicit boundary markers or directives for the agent to ignore potential natural language instructions embedded within the results of its web searches.
- Capability inventory: The agent uses the gathered data for internal scoring and reporting. There are no identified dangerous capabilities (such as
eval,exec, or file-system modifications) that could be triggered by data retrieved from the web. - Sanitization: There is no explicit sanitization or validation logic described for the external content before it is processed by the agent's internal reasoning engine.
Audit Metadata