The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent on how to use various CLI tools for troubleshooting, including Laravel's php artisan, Node.js package managers (npm, pnpm), Rust's cargo, and Docker commands. Notably, references/nuxt-debug.md includes a command to recursively delete directories (rm -rf .nuxt node_modules pnpm-lock.yaml) which is a powerful operation.
[EXTERNAL_DOWNLOADS]: references/laravel-debug.md suggests the installation of several community-standard debugging utilities via Composer, such as beyondcode/laravel-query-detector, itsgoingd/clockwork, and barryvdh/laravel-debugbar.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The skill is designed to ingest and analyze potentially untrusted content such as user-provided error logs, terminal output, and source code files across multiple programming languages.
Boundary markers: There are no explicit instructions or delimiters provided to ensure the agent ignores instructions that might be embedded within the data being debugged.
Capability inventory: The agent has access to capabilities for reading system logs (tail), executing framework commands (php artisan), managing packages (composer, npm), and modifying the file system (rm).
Sanitization: The skill lacks sanitization or validation protocols for external data ingested during the debugging process.

debug