The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection because it processes data from an external, attacker-influenced source (financial market data).
Ingestion points: The script scripts/quote.py accepts a ticker symbol as a command-line argument and retrieves external data via the trading-skills module (fetching from Yahoo Finance).
Boundary markers: There are no delimiters or explicit instructions in the prompt to treat the fetched stock data as untrusted or to ignore embedded instructions.
Capability inventory: The skill executes a Python script using uv run, which has the capability to print output that the agent interprets. No direct file-write or network-send capabilities are present in the provided script code beyond the import.
Sanitization: The input ticker symbol is normalized using .upper(), but the content returned from the external API is not sanitized or validated before being presented to the agent.
[COMMAND_EXECUTION]: The skill uses uv run python scripts/quote.py to execute a local Python script. This is the primary intended mechanism for the skill and is used to process the requested ticker symbol.

stock-quote