The Agent Skills Directory

[COMMAND_EXECUTION]: Analyzes project structure using grep and manages dependencies using npm install or pnpm install based on detected lockfiles.
[EXTERNAL_DOWNLOADS]: Facilitates the download and execution of packages from the npm registry via npx and pnpm dlx.
[EXTERNAL_DOWNLOADS]: Includes a reference to @takazudo/mdx-formatter, which is a tool provided by the author for project formatting.
[PROMPT_INJECTION]: This skill possesses an indirect prompt injection surface as it ingests and processes content from package.json and source code files which could be manipulated by an attacker.
Ingestion points: Reads dependency names, versions, and script definitions from package.json and runs grep on source code.
Boundary markers: None are used to separate untrusted file content from the agent's internal reasoning.
Capability inventory: The skill can modify local files (package.json, hooks) and execute shell commands (npm, pnpm, grep).
Sanitization: No sanitization or verification of the external file content is performed.

dev-npxify