The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves processing data from external URLs, which represents an attack surface for indirect prompt injection. \n
Ingestion points: Untrusted content is ingested via the target URLs provided to the url parameter in various API calls and workflow examples (SKILL.md, api-reference.md). \n
Boundary markers: The provided instructions do not include specific delimiters or guidance for the agent to ignore or isolate instructions found within the scraped content. \n
Capability inventory: The skill enables fetching metadata, performing custom scraping, executing JavaScript (evaluate, function), and generating assets, all of which involve interacting with external network resources. \n
Sanitization: No explicit sanitization or filtering of the target page's content is described in the skill's instructions. \n- [EXTERNAL_DOWNLOADS]: The skill includes instructions to install the microlink.io package via NPM and use the Microlink CLI (SKILL.md). These references target the official resources of the skill's author, microlinkhq. \n- [DYNAMIC_EXECUTION]: The Microlink API provides features for executing JavaScript during the scraping process, as documented in the deep reference (api-reference.md). \n
Browser-side execution: The evaluate parameter allows for running JavaScript in the context of the browser rendering the target page. \n
Server-side execution: The function parameter allows for executing custom JavaScript on the Microlink service's infrastructure with access to the browser session and page content.

microlink-api