The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple external sources, creating a potential surface for indirect prompt injection attacks.
Ingestion points: Financial data and trends are fetched from GitHub (scripts/uptrend_client.py), the Financial Modeling Prep API (scripts/etf_scanner.py), and FINVIZ (scripts/representative_stock_selector.py).
Boundary markers: The Markdown and JSON reports generated by the skill (scripts/report_generator.py) lack explicit delimiters or instructions telling the AI to ignore potential commands embedded in the ingested industry names or stock metadata.
Capability inventory: The skill uses the requests library for network access and performs file system operations to save its findings in a dedicated directory.
Sanitization: While the skill performs rigorous numeric parsing for metrics, it does not sanitize text-based fields like industry names or stock-specific reasons before interpolating them into the final report structure.

theme-detector