The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the npx skills command-line utility to interact with the system environment, specifically for searching, checking, and updating agent capabilities.
[REMOTE_CODE_EXECUTION]: The skill uses the npx skills add command to download and install executable code from remote sources. While it mentions fetching guidelines from Vercel Labs' public repositories as an example, the tool can be used to pull packages from any source accessible via GitHub or the registry.
[PRIVILEGE_ESCALATION]: Instructions provide the agent with the -y flag to bypass confirmation prompts and the -g flag for global installation. This configuration reduces human-in-the-loop oversight for system-altering operations.
[PERSISTENCE]: The skill's primary purpose is the installation of modular packages that persist in the agent's runtime environment across multiple sessions.
[INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect injection by processing untrusted data from an external registry.
Ingestion points: External data enters the agent context via the results of the npx skills find command as described in SKILL.md.
Boundary markers: No delimiters or instructions are present to ensure the agent ignores or sanitizes instructions that might be embedded in search results.
Capability inventory: The agent has the capacity to execute shell commands and perform software installations.
Sanitization: There are no verification or filtering steps defined for the data returned from the external search registry.

find-skills