The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads third-party packages from external registries using the npx package runner. Specifically, it fetches the skills package and various user-selected skill packages.
[REMOTE_CODE_EXECUTION]: The core functionality involves installing and linking third-party 'skills' which contain executable instructions or code. This is performed via the npx skills add command within the scripts/install-skill.sh script.
[COMMAND_EXECUTION]: The skill executes shell commands and CLI tools to search for and install content. It uses npx for package management and performs file system operations (creating directories and symlinks) using mkdir and ln in scripts/install-skill.sh.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection (Category 8) as it processes search results from an external registry.
Ingestion points: Search results returned by the npx skills find command (SKILL.md).
Boundary markers: No explicit delimiters are used when presenting search results to the user or processing them for installation.
Capability inventory: The skill can execute shell scripts, run npx, and modify the project structure through symlinking (scripts/install-skill.sh).
Sanitization: There is no evidence of sanitization or validation of the content retrieved from the external skill registry before it is presented to the user or processed.

find-skills