The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using npx @better-auth/cli for tasks like migrating schemas, generating ORM schemas, and configuring Model Context Protocol (MCP) tools.
[EXTERNAL_DOWNLOADS]: Refers to the official Better Auth website (better-auth.com) and its GitHub repository for source code and documentation. It also facilitates downloading the framework's CLI via npx.
[INDIRECT_PROMPT_INJECTION]: The skill provides a set of instructions that the agent will use to modify and interact with the user's project files.
Ingestion points: The SKILL.md file itself serves as the instruction set for the agent's behavior during integration.
Boundary markers: The content uses standard Markdown headers to structure instructions but lacks explicit guardrail delimiters for user-supplied data handling.
Capability inventory: The instructions involve writing configuration files (auth.ts) and executing shell commands (npx).
Sanitization: No specific sanitization logic is provided in the skill text; it relies on the underlying agent platform's safety protocols.

best-practices