The Agent Skills Directory

[SAFE]: The skill consists exclusively of markdown documentation providing a conceptual framework for AI evaluation (Evals). It does not contain any executable scripts, binary files, or automated commands.
[NO_CODE]: There is no code included in this skill. All logic is presented as text-based instructions or illustrative code templates (e.g., prompt examples and dataset schemas) that do not execute in the agent's environment.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns were found. The skill mentions third-party tools (like LangSmith or Prompt flow) in an informational context but does not download or execute them.

ai-evaluation