The Agent Skills Directory

Strict Grounding Workflow: The skill requires the agent to load specific documentation fragments from the references/ directory, preventing reliance on potentially stale or inaccurate internal model knowledge.
Safe External References: All URLs and repositories mentioned in the skill and its documentation belong to the author's own verified infrastructure or well-known developer platforms, following established security best practices for resource linking.
Intentional Data Retrieval: The instructions for the agent to use grep for searching local markdown files represent a standard and appropriate method for information retrieval within the skill's localized context.
Content Labeling Protocols: Clear instructions are provided to the agent to distinguish and label any information that does not originate directly from the provided playbook fragments, maintaining transparency for the end user.

ax-practitioner