The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands like npx skills find and npx skills add to interact with a skill ecosystem and perform system-level operations.
[REMOTE_CODE_EXECUTION]: The skill instructs the agent to install and execute code from remote GitHub repositories using the npx skills add command. Although it provides guidelines for verifying repository reputation and popularity, this mechanism enables the installation and execution of unverified third-party software.
[EXTERNAL_DOWNLOADS]: The skill fetches data from the skills.sh website and downloads skill packages from GitHub. It specifically references content from trusted organizations like Vercel Labs, Anthropic, and Microsoft.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes untrusted data from search results.
Ingestion points: Output from the npx skills find command is processed by the agent.
Boundary markers: No delimiters are identified to separate search results from instructions.
Capability inventory: The skill has the ability to execute shell commands and install/run remote code.
Sanitization: No automated sanitization is described; the skill relies on the agent's manual verification of metadata.

find-skills