The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The skill ingests untrusted data from external sources, specifically JSON and CSV data from Finviz and FMP APIs via etf_scanner.py and representative_stock_selector.py, as well as a CSV timeseries hosted on GitHub in uptrend_client.py.
Boundary markers: The instructions and scripts do not implement specific boundary markers or 'ignore embedded instructions' warnings for the data being processed.
Capability inventory: The skill's capabilities include data aggregation, mathematical scoring, and generating markdown reports. It does not contain arbitrary shell execution or file-write capabilities using the ingested data.
Sanitization: The skill performs type-checking and numeric parsing (e.g., _safe_float, _parse_market_cap) on all incoming data, which mitigates the risk of malicious payload execution.
[EXTERNAL_DOWNLOADS]:
The skill establishes network connections to download financial data from well-known and established domains including finviz.com, financialmodelingprep.com, and raw.githubusercontent.com. These downloads are documented as functional requirements and target legitimate, well-known services or author-owned repositories.

theme-detector