The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the npx skills command-line interface to perform search, installation, and update operations.
Evidence: Commands like npx skills find, npx skills add, and npx skills update are integrated into the workflow.
[EXTERNAL_DOWNLOADS]: The skill's primary purpose is to facilitate the downloading of external skill packages from sources like GitHub.
Evidence: The npx skills add <package> command fetches modular packages from external repositories to extend agent capabilities.
[REMOTE_CODE_EXECUTION]: By design, the skill installs third-party code packages onto the system. The instructions include a -y flag in the installation command (npx skills add <owner/repo@skill> -g -y), which skips user confirmation prompts during the installation process.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources (the skills.sh website and npx skills find output) to determine which skills to recommend or install. Maliciously crafted skill names or descriptions in the ecosystem could potentially influence the agent's behavior during the discovery and recommendation phase.
Ingestion points: The agent reads search results from npx skills find and the skills.sh leaderboard.
Boundary markers: None specified for external search result processing.
Capability inventory: The agent has the ability to write to the file system and execute code via the skill installation process.
Sanitization: The skill attempts to mitigate risk by instructing the agent to verify metadata like GitHub stars, install counts, and source reputation (e.g., favoring well-known organizations).

find-skills