The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill identifies the project environment and merges configuration rules by reading untrusted local files, which presents a surface for indirect prompt injection.
Ingestion points: Reads project files including package.json, repository lockfiles (pnpm-lock.yaml, yarn.lock, bun.lockb), tsconfig.json, and any existing .dependency-cruiser.* configuration files.
Boundary markers: The instructions do not define specific delimiters or "ignore" instructions for the data read from these files.
Capability inventory: The skill has the capability to execute shell commands via the package manager, write new files (.dependency-cruiser.cjs, README.md, AGENTS.md), and modify the project's package.json scripts.
Sanitization: The skill does not perform specific validation or sanitization on the contents of the lockfiles or existing configuration files before using them to drive installation and configuration steps.
[COMMAND_EXECUTION]: The skill performs shell command execution to manage dependencies and verify the environment setup.
Evidence: Step 2 executes package manager commands to install dependency-cruiser as a devDependency. Step 6 executes the depcruise command multiple times to verify that the newly created rules correctly identify and block restricted imports.

setup-ts-deep-modules