The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run shell commands using the npx skills CLI. This involves executing the skills package directly from the NPM registry.
[EXTERNAL_DOWNLOADS]: The agent is guided to download and install additional skills from remote repositories (e.g., GitHub) using the npx skills add command. This facilitates the addition of external code and workflows to the agent environment.
[PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection (Category 8) by retrieving and processing external data.
Ingestion points: Search results and skill metadata are fetched from the external registry at skills.sh via the npx skills find command.
Boundary markers: No explicit delimiters are used to separate the external metadata from the agent's instructions when presenting options to the user.
Capability inventory: The agent has the capability to execute commands and install further software based on the ingested content.
Sanitization: The skill does not describe any validation or sanitization of the content retrieved from the external registry.

find-skills