The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to scrape and interact with arbitrary web pages, creating a surface for indirect prompt injection where malicious instructions embedded in a website could influence the agent's behavior.
Ingestion points: The scrapeData function in SKILL.md uses page.goto(url) to load external content and page.evaluate() to extract data from the DOM.
Boundary markers: The provided usage examples do not include boundary markers or explicit instructions to ignore content that looks like agent commands.
Capability inventory: The skill includes capabilities to write to the file system (PDF and screenshot generation) and navigate the network via the browser.
Sanitization: The examples do not demonstrate sanitization or validation of the scraped text before returning it to the agent context.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of the puppeteer package from the npm registry.
Evidence: npm install puppeteer is listed in the installation section of SKILL.md. This is a well-known and standard library for browser automation.

puppeteer