The Agent Skills Directory

[COMMAND_EXECUTION] (MEDIUM): The skill frequently executes shell commands including npx, nx, and node. This is central to its functionality but provides a broad attack surface.
[REMOTE_CODE_EXECUTION] (MEDIUM): Step 4 utilizes node -e to dynamically resolve module paths. While intended for discovery, this pattern involves executing strings as JavaScript code. Additionally, running Nx generators (nx generate) executes JavaScript/TypeScript code contained within those generators, which could be malicious if the workspace or plugins are compromised.
[PROMPT_INJECTION] (LOW): The skill metadata contains strong steering instructions such as "INVOKE IMMEDIATELY" and "ALWAYS use this BEFORE", which attempt to override the agent's default tool selection logic.
[EXTERNAL_DOWNLOADS] (LOW): The use of npx can lead to the download and execution of packages from the npm registry if they are not already present in the local cache or node_modules.
[INDIRECT_PROMPT_INJECTION] (LOW): The skill is designed to read and analyze source code and configuration files from the repository (e.g., generators.json, local generator scripts).
Ingestion points: generators.json, generator source code, and existing repository patterns in libs/ or tools/.
Boundary markers: Absent; the skill does not specify delimiters for the data it reads.
Capability inventory: File system read/write, shell command execution (nx, node), and package execution (npx).
Sanitization: None; the skill assumes the repository content and generator logic are trustworthy.

nx-generate