The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill analyzes user-provided decision problems, which creates a surface for instructions to be embedded in input data. \n- Ingestion points: User data is ingested via $ARGUMENTS or conversation context as defined in SKILL.md. \n- Boundary markers: The skill does not implement delimiters or 'ignore' instructions for the processed data. \n- Capability inventory: The agent is instructed to perform a file-write operation to prepare the visualization in SKILL.md. No network or shell capabilities are leveraged. \n- Sanitization: There is no mention of sanitizing or escaping the user-provided data. \n- [DYNAMIC_EXECUTION]: The skill uses an HTML/JavaScript template to present results. \n- Evidence: SKILL.md requires filling markers in assets/template.html and opening the resulting file, which initiates script execution in a browser context.

decision-analysis