The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external APIs, including financial metrics via yahoo_client and trending themes via grok_client. This introduces a potential surface for indirect prompt injection where instructions hidden in stock metadata or social media trends could attempt to influence the agent's behavior.
Ingestion points: Data is retrieved from Yahoo Finance and Grok API (X search) and processed within scripts/run_screen.py for display.
Boundary markers: There are no explicit delimiters or specific instructions for the agent to disregard potential commands embedded within the external data sources.
Capability inventory: The skill is granted restricted execution of Python scripts via the Bash tool and possesses capabilities for network access, file history logging, and knowledge graph updates.
Sanitization: While command-line arguments are structured via argparse, content from external API responses is rendered directly into Markdown tables without specialized sanitization for instruction markers.

screen-stocks