The Agent Skills Directory

[SAFE]: The skill defines a structured methodology for an AI agent to analyze an artifact through multiple independent failure modes. It lacks any functionality for file system modification, network communication, or external command execution.
[NO_CODE]: No scripts, binary files, or software dependencies are included in the skill definition.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze external 'artifacts', which represents a data ingestion surface. 1. Ingestion points: The artifact reviewed in the Workflow section of SKILL.md. 2. Boundary markers: No explicit delimiters or instructions to ignore embedded commands are provided. 3. Capability inventory: The skill does not define or use any tools, scripts, or system-level capabilities. 4. Sanitization: The skill does not include steps to sanitize or filter the input artifact before analysis.

prism