The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to generate and modify Python scripts that may ingest untrusted data from external sources such as APIs or local files.
Ingestion points: User-provided instructions for script generation and potential data processed by the resulting scripts at runtime.
Boundary markers: None explicitly defined in the instructions for isolating external data within the scripts.
Capability inventory: The skill facilitates command execution via uv run, file writes (creating scripts), and network operations through common libraries like requests and httpx shown in examples.
Sanitization: The skill recommends implementing a --dry-run flag for scripts that perform external API calls or file modifications.
[COMMAND_EXECUTION]: The skill uses local shell commands to manage the development lifecycle of Python scripts.
Evidence: Instructions include the use of uv init, uv add, uv run, uv lock, and uv tree for project management.
Evidence: Uses chmod +x to make generated scripts executable, which is standard behavior for CLI tool development.

uv-script