The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to download and run the domain-experts package from the NPM registry. While this appears to be a resource associated with the author's repository, it involves fetching and executing code from an external source at runtime.
[COMMAND_EXECUTION]: The core functionality of the skill relies on executing shell commands via npx to match user requests with available roles.
[REMOTE_CODE_EXECUTION]: The use of npx facilitates remote code execution by pulling the latest version of the specified package from the NPM registry and executing it in the agent's environment.
[PROMPT_INJECTION]: The skill exhibits a risk of indirect prompt injection and command injection. It takes the user's role or task request and interpolates it directly into a shell command: npx domain-experts match "<the role/task in plain language>" --json. If the agent does not properly sanitize the user's input, an attacker could provide a string containing shell metacharacters (e.g., ;, |, `) to execute arbitrary commands on the host system.
Ingestion points: User request string identifying a desired role or expert persona (SKILL.md).
Boundary markers: The instruction suggests wrapping input in double quotes, which provides minimal protection against shell injection.
Capability inventory: Subprocess execution via npx and shell access (SKILL.md).
Sanitization: No instructions are provided to sanitize or validate the user input before it is passed to the shell.

domain-expert-router