Browserious

The page your agent fetched was empty. A page that builds its content in the client returns a script tag and an empty container to a plain HTTP fetch. Browserious gives an agent a real, isolated Chromium session over an API, so it can render that page, interact with it, keep a signed-in profile and return a screenshot.

A browser adds rendered content, interaction, visual verification, persistent signed-in state, and access to a port on your own machine through a tunnel. Sessions are billed by browser time and network traffic; stored data is not metered.

Where to go next

• Browserious in plain text — the full introduction, capabilities and quick start as one text document
• Quick start — install, sign in, and take one screenshot
• Use Browserious from your agent — the agent skill and a setup prompt to paste into your agent
• Pricing — credit packs, what a credit buys, and expiry rules
• API documentation — the full REST reference
• Terms & Conditions, Privacy Policy, Refund Policy

Quick start

Install with pip install browserious, then sign in with browserious login. That is device authorization, so there is no API key to copy. Run browserious session create, which prints a session id, and browserious page create "$SESSION" --url https://example.com, which prints a page id, page-1 for the first page. Both ids are then used by browserious screenshot page "$SESSION" "$PAGE" --output-file result.png.

Use Browserious from your agent

The Browserious skill is one Markdown file that teaches a coding agent the CLI. It is published from the latest CLI release at /skill/SKILL.md; install it where your agent loads skills, for example ~/.claude/skills/browserious/SKILL.md for Claude Code.