The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes user-supplied data in the user_request field to generate summaries and tables, creating a surface for indirect prompt injection. • Ingestion points: src/index.js processes user input via the run function. • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used in the template strings. • Capability inventory: Includes logic for text processing and a subprocess execution capability for validation. • Sanitization: The skill implements a guardRequest check from a shared library to filter inputs.\n- [COMMAND_EXECUTION]: The script scripts/validate_reasoning_artifacts.py utilizes subprocess.run to execute an external Python utility for artifact verification. The target script path is resolved relative to the skill's location, ensuring it remains within the intended shared resource directory.

smart-insightful-friend