The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill instructions direct the agent to dynamically create a Python script ({symbol}_{indicator}_chart.py) based on user-supplied arguments and templates. The script is then executed at runtime to display a Plotly chart.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute the generated Python scripts and perform file system operations, such as creating directory structures for charts.
[INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through its data ingestion and script generation processes.
Ingestion points: User-provided arguments ($ARGUMENTS) and external financial data fetched from OpenAlgo (vendor-provided) or yfinance (well-known service) as specified in SKILL.md.
Boundary markers: None are specified to delimit external data or to warn the agent against executing instructions embedded within the API responses.
Capability inventory: The skill uses Write for script generation and Bash for script execution, providing a pathway for generated code to be run.
Sanitization: The instructions do not include steps for sanitizing or validating user input or external data before it is interpolated into the generated Python code.

indicator-chart