The Agent Skills Directory

[SAFE]: The skill describes a legitimate development process for creating test cases and evaluations within the Prefect ecosystem. All actions described, such as reading documentation, inspecting fixtures, and updating catalogs, are consistent with the skill's stated purpose.- [COMMAND_EXECUTION]: The skill uses the gh CLI and the just command runner to automate tasks. These are standard, well-known developer tools used for repository management and task execution in modern software projects.- [CREDENTIALS_UNSAFE]: The skill includes a specific safety guideline: 'Do not add live credentials, harness-specific argument placeholders, or client-specific tool syntax to the skill.' This instruction proactively mitigates the risk of credential exposure during the evaluation creation process.

implement-eval