macrotrends

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.78). At runtime the skill fetches public Macrotrends HTML pages (e.g., /stocks/charts/{ticker}/{slug}/{endpoint} and the employee table page) and parses embedded var originalData = [...] / table text into JSON that is then printed/returned, so any free text present in those pages (outsider-authored) can become LLM-readable context via the agent’s output.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)