The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses npx ai-agent-skills to perform management tasks such as listing installed skills and checking for available updates.\n- [EXTERNAL_DOWNLOADS]: The check and sync commands trigger network requests to the npm registry and GitHub to verify and download skill updates.\n- [REMOTE_CODE_EXECUTION]: Employs npx to dynamically execute the ai-agent-skills package and uses git clone during synchronization to retrieve remote repository content.\n- [PROMPT_INJECTION]: Detects a vulnerability surface for indirect prompt injection where untrusted data (skill names) from tool output or external registries is interpolated directly into shell commands. Ingestion points: Skill names from list command output or user input. Boundary markers: None present. Capability inventory: Shell command execution via npx and git. Sanitization: No explicit sanitization or validation of skill names is defined before they are passed to the shell.

update-installed-skills