The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes financial and trading information from external sources including SEC EDGAR filings and web search results. This ingestion of untrusted external content represents a surface for indirect prompt injection.
Ingestion points: Data is fetched via web search, web extraction, and browser navigation from sources like SEC.gov and company investor relations pages as fallbacks to MCP data.
Boundary markers: The instructions lack specific boundary markers or instructions for the agent to ignore potentially malicious content within the external data.
Capability inventory: The skill has the ability to write Excel files to disk via openpyxl and execute local Python scripts.
Sanitization: No explicit sanitization or validation of the ingested external content is described.
[COMMAND_EXECUTION]: The skill provides instructions to execute a specific local Python script for workbook recalculation: python /path/to/excel-author/scripts/recalc.py ./out/model.xlsx.
[DYNAMIC_EXECUTION]: The skill uses the openpyxl library to programmatically generate Excel workbooks, including the dynamic creation of spreadsheet formulas from string templates.

comps-analysis