The Agent Skills Directory

[SAFE]: The skill consists of technical reference materials and documentation for the Better Auth library. All content is consistent with its stated purpose of providing an LLM knowledge update for versions 1.2-1.5.- [COMMAND_EXECUTION]: The documentation includes standard CLI commands for the framework, such as npx auth generate and npx auth migrate, which are used for database schema generation and migrations.- [EXTERNAL_DOWNLOADS]: The reference files list legitimate npm packages from the @better-auth and @auth scopes, which are official components of the Better Auth ecosystem.- [INDIRECT_PROMPT_INJECTION]: The skill documents the 'Agent Auth Protocol' and 'OpenAPI adapter', which involve retrieving configuration and capability definitions from external URLs (e.g., discovery endpoints). While this describes a data ingestion surface, it is a functional requirement of the protocol and not a malicious exploit.
Ingestion points: references/agent-auth.md describes methods for fetching OpenAPI specifications and discovery documents.
Boundary markers: The documentation assumes standard integration with the Better Auth library's security features.
Capability inventory: The library provides APIs for authentication management, organization RBAC, and billing.
Sanitization: The framework's internal handling of discovery documents and capability execution provides the necessary validation layers.

better-auth-knowledge-patch