The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the emacsclient command to retrieve theme-specific foreground colors for use in the generated plots.
[REMOTE_CODE_EXECUTION]: It generates Python scripts based on data found in the interaction context and executes them using the uv tool.
[EXTERNAL_DOWNLOADS]: The uv runner is instructed to fetch the matplotlib package from standard public registries as a runtime dependency.
[PROMPT_INJECTION]: The skill processes untrusted data from the conversation history and interpolates it into a Python script template, creating an indirect prompt injection surface.
Ingestion points: Plottable data derived from the current interaction context as described in SKILL.md.
Boundary markers: No explicit delimiters or instructions are provided to prevent embedded data from interfering with the script logic.
Capability inventory: The skill has access to shell execution via the Bash tool, used for executing emacsclient and uv run (SKILL.md).
Sanitization: The instructions do not specify any sanitization or validation of the input data before it is written to the Python script.

matplotlib