The Agent Skills Directory

[SAFE]: The skill consists exclusively of markdown documentation and templates. It does not contain any executable scripts, shell commands, or network operations. It provides a structured methodology for evaluating AI performance and does not interact with the filesystem or external network resources.
[SAFE]: No evidence of credential harvesting, unauthorized network access, or persistence mechanisms was found.
[SAFE]: The inclusion of an MCP server in the frontmatter (@clawfu/mcp-skills) is a standard platform configuration and does not introduce security vulnerabilities in this context.
[SAFE]: The prompt injection examples provided in the "Pressure Testing" section are clearly labeled as test cases for evaluation purposes and do not constitute an attempt to subvert the agent's behavior.

agent-evaluation