The Agent Skills Directory

[SAFE]: The skill is designed to function as a planning and optimization agent for combinatorial interaction testing (CIT). It does not contain any executable scripts, command-line operations, or network-enabled tools, ensuring that its operations are confined to logic processing and plan generation within the agent's environment.
[INDIRECT_PROMPT_INJECTION]: The skill exposes a surface for data ingestion by parsing natural language, YAML, and JSON inputs to extract test axes and values. However, because the skill only generates static plans and does not have the capability to execute the results or communicate externally, the potential for indirect prompt injection to trigger malicious actions is negligible. The documentation also promotes validation and human-in-the-loop review of generated models.
[EXTERNAL_DOWNLOADS]: The skill identifies and references well-known testing tools and academic research from trusted organizations, including NIST (csrc.nist.gov) and Microsoft (github.com/microsoft/pict). These references are provided for user guidance and do not involve automated script downloads or remote code execution patterns.

matrix