browser-harness Setup Guide¶
Skill: browser-harness · Installs: 1,200+ · Source: davidondrej/skills
Browser harness gives AI agents browser automation capabilities - navigate pages, click elements, extract data, fill forms, and take screenshots. Designed for headless operation on VPS and remote Linux servers where GUI browsers aren't available.
Capabilities¶
- Page navigation and URL management
- Element clicking, typing, and form filling
- Content extraction and data scraping
- Screenshot capture for visual verification
- Cookie and session management
Prerequisites¶
Node.js 18+, Playwright or Puppeteer installed on the system.
Installation¶
npx skills add davidondrej/skills --skill browser-harness
Usage¶
Load the skill in your agent session:
/ browser-harness
The skill activates and provides browser automation tools accessible through natural language commands.
Troubleshooting¶
| Issue | Fix |
|---|---|
| Skill not found | Verify npx skills add completed successfully and the skill appears in npx skills list |
| Browser not launching | Ensure Playwright/Puppeteer is installed system-wide |
| Permission errors on VPS | Check that the agent user has access to the browser binary |
Related Skills¶
- vps-server-management - VPS server management and monitoring
- distribute-skill-to-all-agents - Sync skills across multiple agent instances
- Skills Catalog
Integration with CorpusIQ¶
CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.
This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.