The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data from SEC filings and OpenFIGI, creating a theoretical surface for indirect prompt injection attacks.
Ingestion points: The script edgar13f.py fetches XML and JSON content from remote financial data providers.
Boundary markers: The skill does not implement explicit delimiters or instructions to treat the ingested SEC filing content as untrusted data.
Capability inventory: The skill performs filesystem write operations, specifically creating or updating trader/targets.json with processed weights.
Sanitization: Data is parsed structurally using standard Python libraries (json, xml.etree), but the extracted text fields are not sanitized for potential embedded instructions.
[EXTERNAL_DOWNLOADS]: The skill connects to external services to retrieve financial information.
Fetches official corporate filings and information tables from the SEC EDGAR system (sec.gov).
Accesses the OpenFIGI API (api.openfigi.com) to resolve CUSIP identifiers to market tickers.

trader-copycat