The Agent Skills Directory

[SAFE]: The skill implements a strictly template-driven workflow for routine administrative tasks such as appointment scheduling and document collection. This minimizes the risk of generating unvetted or dangerous content.
[SAFE]: It maintains high standards for professional conduct by enforcing a 'supervision-first' policy and requiring the removal of AI-generated labels before external distribution, ensuring human oversight is part of the core process.
[SAFE]: The skill's data access is limited to internal configuration files located in the specific plugin directory (~/.claude/plugins/config/claude-for-legal/), which are used to define pedagogical behavior and contact information. This represents legitimate use of the agent's filesystem for configuration management.

client-letter