The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill guides the user to install several well-known and reputable Node.js packages, including Playwright (Microsoft), JSDOM, and Mozilla Readability. It also involves downloading the Chromium browser binary as part of the Playwright setup process.
[COMMAND_EXECUTION]: The skill uses curl for fetching static web content and node to execute local extraction scripts. These operations are core to the skill's purpose of fetching and parsing web data.
[DYNAMIC_EXECUTION]: The scripts/fetch_js_content.js utility employs Playwright's page.evaluate method to run extraction logic within a headless browser context. This is a standard and necessary technique for scraping JavaScript-rendered websites.
[INDIRECT_PROMPT_INJECTION]: The skill fetches untrusted data from external websites and processes it using the Task agent, which creates a potential surface for indirect prompt injection.
Ingestion points: External URLs fetched via curl or the Playwright-based fetch_js_content.js script (SKILL.md, references/workflows.md).
Boundary markers: The instructions do not currently include explicit boundary markers or directives for the agent to ignore instructions that might be embedded in the fetched content.
Capability inventory: The skill utilizes Bash (for curl/node) and the powerful Task agent capability for content extraction.
Sanitization: The provided scripts use the Mozilla Readability algorithm to strip HTML tags and structure content, which provides structural sanitization but does not filter natural language instructions within the text content.

web-content-fetcher