The Agent Skills Directory

[PROMPT_INJECTION]: The skill ingests untrusted project files and configuration data to detect SDK usage, representing an indirect prompt injection surface. * Ingestion points: Reads package.json and source files (.ts, .js, etc.) using cat, grep, and find commands. * Boundary markers: No explicit delimiters or instructions are provided to the agent to treat file content as untrusted. * Capability inventory: The skill utilizes shell execution for package management and project builds, along with file read/write operations. * Sanitization: No sanitization is performed on external content before it is processed by the agent.
[REMOTE_CODE_EXECUTION]: The skill uses npx @sentry/wizard, which involves downloading and executing a remote tool. This utility is provided by the official vendor (getsentry) and is a standard part of the Sentry migration workflow.
[COMMAND_EXECUTION]: Executes shell commands including npm install, yarn add, pnpm update, npx tsc, and npm run build. These operations are required for the skill's primary purpose of updating dependencies and verifying build integrity.
[EXTERNAL_DOWNLOADS]: References official migration guides and documentation at docs.sentry.io and the source repository at github.com/getsentry. These are trusted resources belonging to the skill author.

sentry-sdk-upgrade