← Back to Skills Marketplace
Nextbrowser
by
Artem Popov
· GitHub ↗
· v1.0.22
· MIT-0
3179
Downloads
17
Stars
3
Active Installs
22
Versions
Install in OpenClaw
/install next-browser
Description
Use Nextbrowser cloud API to spin up cloud browsers for Openclaw to run autonomous browser tasks. Primary use is creating browser sessions with profiles (per...
Usage Guidance
This skill appears to do what it says (manage cloud browser sessions, profiles, proxies). Before installing: 1) Note the SKILL.md reads your OpenClaw config key at skills.entries.next-browser.apiKey but the registry metadata does not list that config path — ask the publisher to declare it. 2) Only provide a Nextbrowser API key you trust and consider creating a dedicated, limited key for this skill. 3) Understand that browser profiles and credential entries (cookies, logins, proxy usernames/passwords) will be stored or transmitted to Nextbrowser — do not upload secrets you wouldn't want the external service to see. 4) If you plan to allow autonomous agent runs, remember the agent can use these browser sessions to act on your accounts; limit scope and monitor activity. 5) Verify Nextbrowser's real domain and privacy/security docs (docs.nextbrowser.com and app.nextbrowser.com) and confirm account retention/purge policies. If the publisher can update the skill metadata to declare the config path (or require an explicit env var) and document the exact data flows, this lowers the risk.
Capability Analysis
Type: OpenClaw Skill
Name: next-browser
Version: 1.0.22
The skill bundle provides a standard integration for the Nextbrowser cloud API, allowing the agent to manage browser profiles and execute autonomous web tasks. The instructions in SKILL.md correctly guide the agent to use the API at app.nextbrowser.com using a user-configured API key, with no evidence of data exfiltration, malicious execution, or unauthorized system access.
Capability Assessment
Purpose & Capability
Name/description match the instructions: the SKILL.md describes spinning up cloud browsers, profiles, proxies, and credential management against nextbrowser API endpoints — these capabilities align with the stated purpose.
Instruction Scope
Instructions only call Nextbrowser endpoints (app.nextbrowser.com) and describe profile, proxy, and credentials operations which are expected. However, the SKILL.md instructs the agent to read an OpenClaw config path (skills.entries.next-browser.apiKey) for the API key — that access was not declared in the registry metadata and should have been.
Install Mechanism
Instruction-only skill with no install steps or downloads — minimal disk/write risk.
Credentials
No env vars or config paths are declared in the registry metadata, yet the runtime docs require an API key stored in OpenClaw config. The skill will cause user account credentials, proxy credentials, and persistent browser profile data to be sent to Nextbrowser (expected for this functionality) — ensure you trust the external service before supplying sensitive data.
Persistence & Privilege
always:false and default autonomous invocation settings. The skill does not request permanent system-wide presence or modify other skills; normal autonomous invocation capability applies.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install next-browser - After installation, invoke the skill by name or use
/next-browser - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.22
No changes detected in code files; minor update to documentation only.
- Added the attached_files field to the Tasks (Subagent) API request in documentation.
- No functional or behavioral changes to the skill itself.
v1.0.21
- Added documentation for using custom proxy servers when creating browser profiles (with required fields and example).
- Clarified usage of proxy modes (`built-in` vs `custom`) in browser profiles.
- Updated example `proxy_settings` in the create profile API to include `custom` mode and all required connection parameters.
- No code changes; documentation only.
v1.0.20
- No file changes detected in this release.
- Content and documentation remain unchanged from the previous version.
v1.0.19
No user-visible changes in this version; no file changes detected.
v1.0.18
- Added documentation for browser_settings.os_type under Profiles, listing supported OS values (windows, linux, macos, android) and default behavior.
- No functional or API changes; update is documentation-only for clarity on operating system selection when creating browser profiles.
v1.0.17
No changes detected in this version.
- No file changes were made for version 1.0.17.
- Functionality and documentation remain unchanged.
v1.0.16
- No changes detected in this version; documentation and implementation remain unchanged.
v1.0.15
next-browser 1.0.15
- Added API endpoints and instructions to start and stop browser instances for profiles.
- Now documents the requirement to start a browser instance after creating or modifying a profile, and to poll for `active` status before proceeding.
- Includes sample cURL commands for starting, stopping, and checking status of browser profiles.
- Clarified that tasks require an active (running) browser profile, and will fail otherwise.
v1.0.14
- Added support for the send_email_notification option to browser task requests.
- Updated task API documentation to describe the send_email_notification field and its effect.
- No other functional or documentation changes.
v1.0.13
Version 1.0.13
- No file or documentation changes detected in this release.
- No updates to features, endpoints, or usage instructions.
v1.0.12
- Changed HTTP method for stopping a chat session/task from POST to PUT in the "Tasks (Subagent)" section.
- No other changes detected.
v1.0.11
- Added documentation for stopping chat sessions and tasks via POST /api/v1/chat/sessions/<chat-session-id>/stop.
- No other content changes detected.
v1.0.10
- Added explicit instruction to not validate Nextbrowser API key formats; keys should be accepted as provided by the user.
- Updated API key setup instructions, including a generic placeholder and warning about possible formats.
- Clarified error handling: authentication errors should only prompt user action after a failed API response.
- No changes to file structure; content update only.
v1.0.9
- Improved the Tasks endpoint documentation with details for polling task completion steps, including an example using from_step and explanation of new response fields.
- Clarified the meaning of status, steps, step_number, and total_steps in task polling.
- Added instructions on using optional query parameters when retrieving task status.
- No breaking changes to existing behavior or API usage.
v1.0.8
Version 1.0.8 – No functional changes.
- No file changes detected; documentation and functionality remain the same as the previous version.
v1.0.7
- Reordered sections: Credentials Manager is now listed before Profiles.
- Profiles endpoint updated—browser profile creation now supports a credentials field for associating credentials with a new profile.
- The subagent Tasks endpoint documentation was updated to remove credentials from example payloads.
- No code changes detected; this is a documentation update only.
v1.0.6
- Added documentation for the Locations API endpoints, including how to list countries, regions, cities, and ISPs for configuring proxies and browser geolocation.
- Incremented Tasks (Subagent) section to section 4 to accommodate the new Locations section.
- No other significant changes.
v1.0.5
- Added documentation for the new Credentials Manager feature, including usage example.
- Updated task automation example to demonstrate using credentials for login tasks.
- Expanded description to highlight new features: residential proxy, browser stealth, and CAPTCHA solving.
- Revised documentation for clarity and to reflect current task options and use cases.
v1.0.4
- Renamed documentation file from SKILL.MD to SKILL.md to use correct filename casing.
- No changes to feature functionality or API usage.
v1.0.2
- Fixed documentation of task status values: changed "completed" to "finished" and removed "stopped".
- Minor documentation improvements for accuracy and clarity.
Metadata
Frequently Asked Questions
What is Nextbrowser?
Use Nextbrowser cloud API to spin up cloud browsers for Openclaw to run autonomous browser tasks. Primary use is creating browser sessions with profiles (per... It is an AI Agent Skill for Claude Code / OpenClaw, with 3179 downloads so far.
How do I install Nextbrowser?
Run "/install next-browser" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Nextbrowser free?
Yes, Nextbrowser is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Nextbrowser support?
Nextbrowser is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Nextbrowser?
It is built and maintained by Artem Popov (@highxshell); the current version is v1.0.22.
More Skills