The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to use npx to download and execute the @useclawlink/cli package from the NPM registry, which is the official distribution channel for the vendor's integration tools.
[COMMAND_EXECUTION]: All Vercel operations are performed through shell commands invoking the ClawLink CLI for authentication and resource management.
[PROMPT_INJECTION]: The skill has an indirect prompt injection attack surface.
Ingestion points: The agent retrieves data from the Vercel API, including project names, domain lists, and deployment details, through actions such as vercel_get_projects and vercel_get_deployments (SKILL.md).
Boundary markers: There are no specific delimiters or instructions provided to the agent to treat data from the Vercel API as untrusted or to ignore embedded instructions.
Capability inventory: The skill possesses the ability to perform administrative writes, including vercel_add_environment_variable, vercel_create_project2, and vercel_create_new_deployment (SKILL.md).
Sanitization: No validation or sanitization of the external data is performed before it enters the agent's context.

vercel