The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands for project discovery and verification.
The Phase 1: Detect section in SKILL.md uses grep to scan Gemfile and other source files for framework markers, background job processors, and competitor monitoring tools.
The verification section suggests running npx @spotlightjs/spotlight to launch a local development sidecar.
These commands are standard for project analysis and developer workflows.
[EXTERNAL_DOWNLOADS]: The skill guides the user to install external dependencies.
Recommends installation of official Sentry gems (e.g., sentry-ruby, sentry-rails, sentry-sidekiq) via bundle install.
Recommends the usage of @spotlightjs/spotlight via npx for local event inspection.
These resources are official packages from the vendor and are standard for the SDK's functionality.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection via project metadata files.
Ingestion points: The skill reads content from Gemfile, package.json, and source code files in app/, lib/, and config/ using grep and cat in SKILL.md.
Boundary markers: None present; the skill treats the output of these discovery commands as reliable indicators for routing logic.
Capability inventory: The skill can suggest command executions (installation) and configuration changes based on the data found.
Sanitization: None; it performs simple string matching to detect project state.

sentry-ruby-sdk