Skip to content
Back to search

browserbase

MarketplaceWorks with Claude CodeGitHub

Browserbase's official collection of agent skills to access the web.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/browserbase/skills with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/browserbase/skills'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/browserbase-skills/badge.svg)](/listings/browserbase-skills)

HTML

<a href="/listings/browserbase-skills"><img src="/api/listings/browserbase-skills/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

To install the skill to popular coding agents:

$ npx skills add browserbase/skills

Claude Code

On Claude Code, to add the marketplace, simply run:

/plugin marketplace add browserbase/skills

Then install the plugin:

/plugin install browse@browserbase

If you prefer the manual interface: 1. On Claude Code, type `/plugin` 2. Select option `3. Add marketplace` 3. Enter the marketplace source: `browserbase/skills` 4. Press enter to select the `browse` plugin 5. Hit enter again to `Install now` 6. **Restart Claude Code** for changes to take effect

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 103 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. {
  2. "name": "browserbase",
  3. "owner": {
  4. "name": "Browserbase",
  5. "email": "support@browserbase.com"
  6. },
  7. "metadata": {
  8. "description": "Browser automation and Browserbase CLI skills",
  9. "version": "1.0.0"
  10. },
  11. "plugins": [
  12. {
  13. "name": "browse",
  14. "source": "./",
  15. "description": "Automate web browser interactions using natural language. Use when the user asks to browse websites, navigate web pages, extract data from websites, take screenshots, fill forms, click buttons, or interact with web applications.",
  16. "version": "0.0.1",
  17. "author": {
  18. "name": "Browserbase"
  19. },
  20. "category": "automation",
  21. "keywords": ["browser", "automation", "web-scraping", "stagehand", "screenshots"],
  22. "strict": false,
  23. "skills": [
  24. "./skills/browser"
  25. ]
  26. },
  27. {
  28. "name": "functions",
  29. "source": "./",
  30. "description": "Deploy serverless browser automation to the cloud using Browserbase Functions. Use when the user wants to deploy automation to run on a schedule, needs a webhook endpoint for browser automation, wants to run automation in the cloud instead of locally, or asks about Browserbase Functions.",
  31. "version": "0.0.1",
  32. "author": {
  33. "name": "Browserbase"
  34. },
  35. "category": "automation",
  36. "keywords": ["serverless", "functions", "deployment", "cloud", "webhook", "scheduled-automation"],
  37. "strict": false,
  38. "skills": [
  39. "./skills/functions"
  40. ]
  41. },
  42. {
  43. "name": "browser-trace",
  44. "source": "./",
  45. "description": "Capture a full DevTools-protocol trace (CDP firehose, screenshots, DOM dumps) alongside any browser automation, then bisect the stream into per-page searchable buckets. Use when the user wants to debug a failed run, audit network/console/DOM activity, attach a trace to an in-flight Browserbase session, or feed structured per-page summaries back into an agent loop.",
  46. "version": "0.0.1",
  47. "author": {
  48. "name": "Browserbase"
  49. },
  50. "category": "automation",
  51. "keywords": ["debugging", "trace", "observability", "cdp", "devtools-protocol", "session-debugging", "network-audit"],
  52. "strict": false,
  53. "skills": [
  54. "./skills/browser-trace"
  55. ]
  56. },
  57. {
  58. "name": "safe-browser",
  59. "source": "./",
  60. "description": "Build local constrained-browser agents with a safe_browser tool that owns CDP, enforces a domain allowlist with Fetch interception, and lets a runtime Claude Agent SDK agent complete browsing tasks without raw browser, shell, or CDP access.",
  61. "version": "0.0.1",
  62. "author": {
  63. "name": "Browserbase"
  64. },
  65. "category": "security",
  66. "keywords": ["safe-browser", "domain-policy", "allowlist", "cdp", "fetch-interception", "claude-agent-sdk", "browser-security", "prompt-injection"],
  67. "strict": false,
  68. "skills": [
  69. "./skills/safe-browser"
  70. ]
  71. },
  72. {
  73. "name": "webmcp-gen",
  74. "source": "./",
  75. "description": "Author, compile, and validate site-specific WebMCP init scripts from a target URL and desired tool capability.",
  76. "version": "0.0.1",
  77. "author": {
  78. "name": "Browserbase"
  79. },
  80. "category": "automation",
  81. "keywords": ["webmcp", "stagehand", "browser", "automation", "tool-generation", "chrome-devtools"],
  82. "strict": false,
  83. "skills": [
  84. "./skills/webmcp-gen"
  85. ]
  86. },
  87. {
  88. "name": "add-webmcp",
  89. "source": "./",
  90. "description": "Analyze an existing web application, add first-party WebMCP tools backed by its routes, forms, actions, and schemas, and validate them with Stagehand v4.",
  91. "version": "0.0.1",
  92. "author": {
  93. "name": "Browserbase"
  94. },
  95. "category": "automation",
  96. "keywords": ["webmcp", "stagehand-v4", "code-analysis", "tool-generation", "agent-ready"],
  97. "strict": false,
  98. "skills": [
  99. "./skills/add-webmcp"
  100. ]
  101. }
  102. ]
  103. }
README.md · README · 95 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # Browserbase Skills
  2. A set of skills for enabling **[Claude Code](https://docs.claude.com/en/docs/claude-code/overview)** to work with Browserbase through browser automation and the official `browse` CLI.
  3. ## Skills
  4. This plugin includes the following skills (see `skills/` for details):
  5. | Skill | Description |
  6. |-------|-------------|
  7. | [browser](skills/browser/SKILL.md) | Automate web browser interactions via CLI commands — supports remote Browserbase sessions with Browserbase Identity, Verified browsers, CAPTCHA solving, and residential proxies |
  8. | [functions](skills/functions/SKILL.md) | Deploy serverless browser automation to Browserbase cloud using the `browse` CLI |
  9. | [browser-trace](skills/browser-trace/SKILL.md) | Capture a full DevTools-protocol trace (CDP firehose, screenshots, DOM dumps) alongside any browser automation, then bisect the stream into per-page searchable buckets |
  10. | [browser-to-api](skills/browser-to-api/SKILL.md) | Turn a website's observable HTTP traffic into a best-effort OpenAPI 3.1 spec by analyzing a `browser-trace` capture |
  11. | [autobrowse](skills/autobrowse/SKILL.md) | Self-improving browser automation — iteratively runs a browsing task, reads the trace, and improves the navigation skill until it reliably passes |
  12. | [optimize-agent-prompt](skills/optimize-agent-prompt/SKILL.md) | Optimize Browserbase Agent API system prompts through repeated runs, Agent message traces, session logs, scoring, and unchanged confirmation runs |
  13. | [verified-eval](skills/verified-eval/SKILL.md) | Compare the same browser task in Standard and Verified sessions with an existing flow or a Stagehand v4 agent, then report task scores and timing |
  14. | [safe-browser](skills/safe-browser/SKILL.md) | Build local Claude Agent SDK browser agents whose only browser capability is a CDP-gated `safe_browser` tool with domain allowlist enforcement |
  15. | [webmcp-gen](skills/webmcp-gen/SKILL.md) | Author, compile, and validate site-specific WebMCP init scripts with the Stagehand WebMCP runtime |
  16. | [add-webmcp](skills/add-webmcp/SKILL.md) | Analyze an existing web app, add first-party WebMCP tools backed by its routes, forms, actions, and schemas, and validate them through Stagehand |
  17. | [cookie-sync](skills/cookie-sync/SKILL.md) | Sync cookies from local Chrome to a Browserbase persistent context so the browse CLI can access authenticated sites |
  18. | [fetch](skills/fetch/SKILL.md) | Fetch HTML or JSON from static pages without a browser session — inspect status codes, headers, follow redirects |
  19. | [search](skills/search/SKILL.md) | Search the web and return structured results (titles, URLs, metadata) without a browser session |
  20. | [ui-test](skills/ui-test/SKILL.md) | AI-powered adversarial UI testing — analyzes git diffs to test changes, or explores the full app to find bugs |
  21. | [browser-use-to-stagehand](skills/browser-use-to-stagehand/SKILL.md) | Migrate browser-use (Python) automation to Stagehand v3 (TypeScript) on Browserbase — maps features and picks the right determinism level per step |
  22. | [agent-experience](skills/agent-experience/SKILL.md) | Audit how agent-friendly a product, SDK, or docs site is — drops Claude subagents at it with tiny prompts, captures their traces, and scores setup friction, speed, error recovery, and doc quality |
  23. | [company-research](skills/company-research/SKILL.md) | Discover target companies matching your ICP using the Browserbase Search API, deep-research each one, and score fit into a research report and CSV |
  24. | [event-prospecting](skills/event-prospecting/SKILL.md) | Extract speakers from a conference page, filter their companies against your ICP, and deep-research the best-fit people into a person-first prospecting report |
  25. | [competitor-analysis](skills/competitor-analysis/SKILL.md) | Auto-discover a company's competitors via the Browserbase Search API, deep-research each across marketing, signal, benchmark, and strategic-diff lanes, and compile a browsable HTML report with an overview, per-competitor deep dives, a feature/pricing matrix, and a mentions feed |
  26. To install only the Verified eval skill, run `npx skills add browserbase/skills --skill verified-eval`.
  27. ## Installation
  28. To install the skill to popular coding agents:
  29. ```bash
  30. $ npx skills add browserbase/skills
  31. ```
  32. ### Claude Code
  33. On Claude Code, to add the marketplace, simply run:
  34. ```bash
  35. /plugin marketplace add browserbase/skills
  36. ```
  37. Then install the plugin:
  38. ```bash
  39. /plugin install browse@browserbase
  40. ```
  41. If you prefer the manual interface:
  42. 1. On Claude Code, type `/plugin`
  43. 2. Select option `3. Add marketplace`
  44. 3. Enter the marketplace source: `browserbase/skills`
  45. 4. Press enter to select the `browse` plugin
  46. 5. Hit enter again to `Install now`
  47. 6. **Restart Claude Code** for changes to take effect
  48. ## Usage
  49. Once installed, you can ask Claude to browse or use the Browserbase CLI:
  50. - *"Go to Hacker News, get the top post comments, and summarize them "*
  51. - *"QA test http://localhost:3000 and fix any bugs you encounter"*
  52. - *"Order me a pizza, you're already signed in on Doordash"*
  53. - *"Use `browse` to list my Browserbase projects and show the output as JSON"*
  54. - *"Initialize a new Browserbase Function with `browse functions init` and explain the next commands"*
  55. - *"Use safe-browser to build a Hacker News scraper that only stays on the main site"*
  56. Claude will handle the rest.
  57. For local and localhost work, pass `--local` on the first browser command (for example, `browse open http://localhost:3000 --local`) to start a clean isolated browser. Use `--auto-connect` when the agent should reuse your existing local Chrome session, cookies, or login state.
  58. ## Troubleshooting
  59. ### Chrome not found
  60. Install Chrome for your platform:
  61. - **macOS** or **Windows**: https://www.google.com/chrome/
  62. - **Linux**: `sudo apt install google-chrome-stable`
  63. ### Profile refresh
  64. To refresh cookies from your main Chrome profile:
  65. ```bash
  66. rm -rf .chrome-profile
  67. ```
  68. ## Resources
  69. - [Stagehand Documentation](https://github.com/browserbase/stagehand)
  70. - [Claude Code Skills](https://support.claude.com/en/articles/12512176-what-are-skills)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.