The Agent Skills Directory

[SAFE]: The skill employs standard, read-only shell commands (such as cat, ls, and find) to detect the project environment, including router types, dependencies, and configuration files. This is a legitimate development workflow for an automated setup guide.
[SAFE]: External dependencies and installation commands (npm install @sentry/nextjs, npx @sentry/wizard) target official packages from Sentry, a well-known and reputable technology service. The skill explicitly instructs the user to execute the interactive wizard themselves, ensuring user oversight.
[SAFE]: Security best practices are encouraged for managing sensitive data. For example, the skill recommends using environment variables for the SENTRY_AUTH_TOKEN and ensures that sensitive files like .env.sentry-build-plugin are added to .gitignore.
[SAFE]: The identification of backend services in adjacent directories (../backend, ../server) is performed using minimal reads (head -3) for the purpose of configuring distributed tracing, which is a core feature of the monitoring SDK being set up.
[SAFE]: AI monitoring instructions include explicit warnings about PII (Personally Identifiable Information) and provide opt-in configuration patterns for capturing prompts and completions.

sentry-nextjs-sdk