The Agent Skills Directory

[COMMAND_EXECUTION]: The skill employs standard shell commands such as find, grep, ls, and cat to analyze project files (e.g., .csproj, package.json) and identify framework types and existing Sentry configurations. This is a legitimate discovery process intended for environmental setup.\n- [EXTERNAL_DOWNLOADS]: Recommends downloading the Sentry SDK via the official NuGet registry and running the Sentry setup wizard via npx. These operations target well-known package managers and trusted vendor-owned resources.\n- [REMOTE_CODE_EXECUTION]: Provides instructions to execute the official Sentry setup wizard using npx @sentry/wizard@latest. This is a standard installation method for the Sentry SDK and originates from a well-known service.\n- [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection by reading local project configuration files (Ingestion points: .csproj, package.json, Program.cs). The instructions do not define specific delimiters for this data (Boundary markers: Absent), and the agent has the capability to execute package managers or modify project files (Capability inventory: npx, dotnet, file-write). No specific sanitization of these file contents is performed (Sanitization: Absent). The risk is categorized as low as it involves standard project files within the user's environment.

sentry-dotnet-sdk