The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use the terminal to perform environment checks (e.g., pip show) and to execute the prototype scripts it creates.\n- [DYNAMIC_EXECUTION]: The core purpose involves the dynamic generation of source code files (Python, Node.js) which are then executed within the agent's environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill's research phase involves ingesting data from arbitrary web URLs and local files, which could contain instructions that influence the agent's subsequent coding and execution actions.\n
Ingestion points: External content fetched via web_extract and local library content read via read_file.\n
Boundary markers: The skill does not define specific delimiters or instructions to ignore potential commands within the researched documentation.\n
Capability inventory: The agent has access to terminal for shell execution, write_file for creating persistent scripts, and delegate_task for sub-tasking.\n
Sanitization: There is no logic provided to sanitize or validate the content retrieved from the web before it is used to guide the implementation of spikes.

spike