web-research

name	web_research
description	Research a topic in a live browser using PageState-based navigation, extraction, and evidence synthesis.
official	true
version	0.1.0

Web Research Skill

Use this skill when the user wants live information gathered from websites instead of a static codebase or document search.

Goal

Produce a short, evidence-backed answer by navigating pages, reading structured browser_get_page output, and extracting only the content needed to answer the user's question.

Preferred Tool Sequence

1. Use browser_navigate to open the destination or a search engine.
2. Use browser_get_page after every meaningful navigation to inspect the current PageState.
3. Use browser_click and browser_type with backend_node_id when available; otherwise use element_id.
4. Use browser_press_key after typing into search boxes or forms.
5. Use browser_wait when the page is still loading or a result list is expected.
6. Use browser_extract_content or browser_get_text only after landing on the right page.

Operating Rules

1. Prefer pages that already show the answer directly; do not open unnecessary tabs.
2. Re-read browser_get_page after clicks, form submits, or infinite-scroll changes.
3. If the page state warns about login, captcha, or blocked flows, stop and tell the user exactly what is needed.
4. Summaries must distinguish observed facts from inference.
5. Cite the page title and URL in the final natural-language answer.

Minimal Demo Flow

1. browser_navigate to the target page or search engine.
2. browser_get_page to find the primary result.
3. browser_click the result using backend_node_id.
4. browser_extract_content to gather the answer.
5. Return a concise answer with the source URL.