Web Browse Skill
Use the browser tool to navigate web pages as a real user.
Mandatory execution rule
Use this skill only when normal brave-search is insufficient: login flows, dynamic pages, forms, buttons, CAPTCHAs, dashboards, or explicit requests to open/navigate a site. For ordinary web search, use brave-search first. Do not tell the user web access is unavailable just because browser interaction is not needed.
Setup
Browser is already configured on Argo:
- Chromium at
/usr/bin/chromium (no sandbox, host mode)
- CDP port: 18800
Browser Tool Quick Reference
| Action |
When to use |
browser(action="open", url="...") |
Navigate to URL |
browser(action="snapshot", targetId="...") |
Get page structure (aria refs) |
browser(action="screenshot", targetId="...") |
Visual screenshot |
browser(action="act", targetId="...", request={kind:"click", ref:"..."}) |
Click element |
browser(action="act", targetId="...", request={kind:"type", ref:"...", text:"..."}) |
Type in input |
browser(action="act", targetId="...", request={kind:"press", key:"Enter"}) |
Press key |
Workflow
- Open the URL with
browser(action="open", target="host", url="...")
- Snapshot to see page structure and get element refs
- Act to interact (click, type, scroll, hover)
- Repeat until task is done
- Close the browser when done
Common Patterns
Login to a site
open → snapshot → find login fields → type credentials → click submit → verify
Extract data from a page
open → snapshot → identify data elements → act to expand/scroll if needed → snapshot → extract
Click through a flow
open → snapshot → click button → wait for navigation → snapshot → click next → ...
Tips
- Use
compact=true in snapshot for fewer nodes
- Use
refs="aria" to get stable aria refs for act targets
- If a page doesn't load, try adding
timeMs=2000 to act for wait
- For pages requiring user interaction, use
screenshot to see what the agent sees
- Headless mode: add
headless=true to open if you don't want visible browser
- Keep browser open between related actions to avoid re-authenticating
1---2name: web-browse3description: Navigate websites and interact with web pages as a human user. Use when needing to browse, scrape, extract data from, or interact with websites that require real browser behavior (login flows, dynamic content, CAPTCHAs, bot detection, etc.). Triggers on phrases like "browse to", "go to website", "open webpage", "check the page", "extract from", "login to", "navigate to", or any task requiring real web interaction with a browser.4---56# Web Browse Skill78Use the `browser` tool to navigate web pages as a real user.910## Mandatory execution rule1112Use this skill only when normal `brave-search` is insufficient: login flows, dynamic pages, forms, buttons, CAPTCHAs, dashboards, or explicit requests to open/navigate a site. For ordinary web search, use `brave-search` first. Do not tell the user web access is unavailable just because browser interaction is not needed.1314## Setup1516Browser is already configured on Argo:17- Chromium at `/usr/bin/chromium` (no sandbox, host mode)18- CDP port: 188001920## Browser Tool Quick Reference2122| Action | When to use |23|--------|-------------|24| `browser(action="open", url="...")` | Navigate to URL |25| `browser(action="snapshot", targetId="...")` | Get page structure (aria refs) |26| `browser(action="screenshot", targetId="...")` | Visual screenshot |27| `browser(action="act", targetId="...", request={kind:"click", ref:"..."})` | Click element |28| `browser(action="act", targetId="...", request={kind:"type", ref:"...", text:"..."})` | Type in input |29| `browser(action="act", targetId="...", request={kind:"press", key:"Enter"})` | Press key |3031## Workflow32331. **Open** the URL with `browser(action="open", target="host", url="...")`342. **Snapshot** to see page structure and get element refs353. **Act** to interact (click, type, scroll, hover)364. **Repeat** until task is done375. **Close** the browser when done3839## Common Patterns4041### Login to a site42```43open → snapshot → find login fields → type credentials → click submit → verify44```4546### Extract data from a page47```48open → snapshot → identify data elements → act to expand/scroll if needed → snapshot → extract49```5051### Click through a flow52```53open → snapshot → click button → wait for navigation → snapshot → click next → ...54```5556## Tips5758- Use `compact=true` in snapshot for fewer nodes59- Use `refs="aria"` to get stable aria refs for act targets60- If a page doesn't load, try adding `timeMs=2000` to act for wait61- For pages requiring user interaction, use `screenshot` to see what the agent sees62- Headless mode: add `headless=true` to open if you don't want visible browser63- Keep browser open between related actions to avoid re-authenticating
LFTPadilla