The Agent Skills Directory

[SAFE] (SAFE): No malicious patterns or security vulnerabilities were detected. The skill consists of legitimate documentation and boilerplate code for the Vercel AI SDK.\n- [EXTERNAL_DOWNLOADS] (SAFE): The skill references and provides installation instructions for well-known, trusted packages from verified organizations (Vercel, OpenAI, Anthropic, Google, Cloudflare) via standard package managers.\n- [COMMAND_EXECUTION] (SAFE): The check-versions.sh script uses benign npm list and npm view commands for version management, presenting no security risk to the user's environment.\n- [CREDENTIALS_UNSAFE] (SAFE): All provided templates correctly demonstrate the use of environment variables for handling API keys, following security best practices to prevent credential exposure.

ai-sdk-core