The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs extensive shell command execution using the Nx CLI to manage workspace projects.
Commands include npx nx list, npx nx g (generate), nx format, and nx run-many for verification.
These operations are fundamental to the skill's purpose of project scaffolding and management.
[DYNAMIC_EXECUTION]: The skill uses node -e to execute a JavaScript one-liner for resolving the filesystem paths of plugin configuration files.
Evidence: node -e "console.log(require.resolve('@nx/<plugin>/generators.json'));" in SKILL.md.
This is used as a discovery mechanism to find source code within node_modules.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a bridge between user requests and shell execution, creating a surface for indirect prompt injection.
Ingestion points: User-supplied project names, directory paths, and framework preferences (triggers like "create a ... app").
Boundary markers: None explicitly defined in the instructions to separate user input from the shell command template.
Capability inventory: Subprocess execution via nx generate, nx run-many, and node -e (found in SKILL.md).
Sanitization: No specific sanitization or escaping instructions are provided for user-supplied arguments before they are passed to the CLI.
[EXTERNAL_DOWNLOADS]: The skill utilizes npx, which may download and execute packages from the npm registry if the required Nx tools or plugins are not already cached locally.

nx-generate