The Agent Skills Directory

Command Execution: The skill executes CLI commands such as vercel logs and vercel env ls to gather diagnostic information. While running commands is a sensitive capability, these tools are used here to provide visibility into application behavior and environment status.\n- Sensitive File Access: The skill reads .env.local and other project files to verify that necessary environment variables are present and correctly configured. This allows the agent to identify missing or incorrect settings that could prevent the application from functioning correctly.\n- Indirect Prompt Injection Surface: The skill analyzes recently edited source code and project configuration files to infer the current development task. While this provides necessary context for verification, it is a point where external content enters the agent's context.\n
Ingestion points: Project source files, package.json, and environment files.\n
Boundary markers: No specific boundary markers are defined in the instructions.\n
Capability inventory: Execution of CLI diagnostic commands and file system read access.\n
Sanitization: No explicit sanitization or filtering of ingested file content is described.

verification