The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the @jackwener/opencli package from the NPM registry and suggests downloading a browser extension from its official GitHub repository to enable full functionality.\n- [COMMAND_EXECUTION]: The skill executes shell commands using the opencli tool for data retrieval. It also contains a shell command executed at skill load time (Dynamic Context Injection) to verify if the tool is installed and functioning correctly.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it retrieves and summarizes content from a wide range of untrusted external websites.\n
Ingestion points: Data returned by opencli from over 90 external sites including social media, news, and research databases.\n
Boundary markers: The instructions lack explicit delimiters or safety warnings to prevent the agent from being influenced by instructions embedded in the fetched web content.\n
Capability inventory: The skill enables shell execution for the opencli tool suite.\n
Sanitization: There is no evidence of sanitization or filtering of external content before it is processed by the agent.

opencli-reader