agent-labs-dev/fastbrowse/docs

Use fastbrowse with your agent

The fastbrowse skill lets an agent that supports Agent Skills delegate a website task to fastbrowse and interpret its status and citations. It prefers a separate headless local Chrome, and drives another browser only through a DevTools endpoint the user names. It uses an MCP server configured for that browser choice, or the CLI. Cloud browsing is available when requested or local browsing is unavailable. Agents without skill support can use the CLI or MCP server directly. Installing the skill installs instructions; the browser and model credentials still need configuration.

Install

From a checkout of this repository, install the skill for your agent with the Skills CLI:

npx skills add ./ --skill fastbrowse

The same installer can fetch the skill from GitHub without a checkout:

npx skills add agent-labs-dev/fastbrowse --skill fastbrowse

Choose your agent when the installer prompts. The installer uses project scope by default. Add --global for all projects, or --agent codex claude-code to select those two agents explicitly. Review the skill before installing it. For a manual install, copy skills/fastbrowse into .agents/skills/fastbrowse for Codex or .claude/skills/fastbrowse for Claude Code in the project where you want to browse. Keep its references directory beside SKILL.md. Personal installs use ~/.agents/skills/fastbrowse or ~/.claude/skills/fastbrowse. Start a new agent session after installation.

Configure and use

Install uv for the CLI path. Set TYPESAFE_API_KEY for Jev, direct TypeSafe, which is the recommended route, and OPENROUTER_API_KEY for the LLM. If no direct Jev key is available, OpenRouter also serves Jev text, but it is not recommended and never receives Jev image input. AI_GATEWAY_API_KEY is the LLM fallback, and on its own also serves Jev as a last resort rather than the recommended Jev route. Set keys in the calling process environment or a local .env, outside the conversation. Local Chrome is the skill default and needs only the model key. Add BROWSER_USE_API_KEY for cloud browsing. An existing DevTools connection reuses your Chrome; --local without a connection starts a separate browser. The CLI guide covers browser profiles, secrets, and budgets. For MCP, use the server setup instead; credentials belong to the server's environment.

In Codex:

Use $fastbrowse to find the top story on https://news.ycombinator.com/ and return its title with a citation.

In Claude Code:

/fastbrowse Find the top story on https://news.ycombinator.com/ and return its title with a citation.

The skill also supports automatic selection from its description. It uses the CLI's uncapped resource defaults unless you supply limits or an existing task budget requires them. MCP server ceilings still apply. Cloud browser charges are added when the browser stops, separately from any model-spend cap.

For a form, include its URL and the values to fill. State whether the agent should stop before submission or submit the specified content. The skill does not authorize writes on its own. A result of complete means the task was verified; other statuses describe the missing input, authorization, evidence, or access.

Skill design and testing

The skill uses the shared Agent Skills format. Its description names the browsing tasks it handles, and its main file links to browser access details when needed. This follows the Codex skill guidance and Claude authoring guidance.

Test a skill change in a fresh agent session with real model calls and a browser. Check the browser's result as well as the calling agent's answer: a cited read, interactive navigation, a form stopped before submission, and an explicitly authorized submission. Grade the form against the server's receipt or records, not the agent's description. Also exercise missing configuration and exhausted budgets so unfinished runs remain unfinished. Run the documented commands from an installed copy outside this checkout, where repository docs cannot fill gaps in the skill. Test automatic selection separately from explicit invocation. These are behavioral checks, not assertions about the skill's wording.