docs: update browser automation section with all tools in tested preference order#162
Conversation
…erence order Add Agent Browser, Dev-Browser, Playwriter, and Sweet Cookie which were missing. Reorder all tools by preference from OpenCode session testing: Agent Browser (default, snapshot+ref), Dev-Browser (stateful), Playwriter (existing sessions), Sweet Cookie (cookie extraction), Stagehand (NLP), Crawl4AI (scraping), Playwright, Chrome DevTools, Cloudflare Browser. Include rationale paragraph explaining why each tool ranks where it does based on reliability, cost, and use-case fit for AI agents.
Summary of ChangesHello @marcusquinn, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request significantly enhances the documentation for browser automation tools by incorporating additional options and re-evaluating their presentation. The primary goal is to provide users with a more complete and prioritized list, reflecting practical testing insights and offering clear guidance on tool selection based on specific use cases. Highlights
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
|
Warning Rate limit exceeded
⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. Note Other AI code review bot(s) detectedCodeRabbit has detected other AI code review bot(s) in this pull request and will avoid duplicating their findings in the review comments. This may lead to a less comprehensive review. WalkthroughUpdated the README.md's Browser Automation section with a preference-based tool ordering and explanatory rationale. Introduced new tools (Agent Browser, Dev-Browser, Playwriter, Sweet Cookie) and reordered existing ones with expanded descriptions based on reliability, determinism, and latency considerations. Changes
Estimated code review effort🎯 1 (Trivial) | ⏱️ ~3 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
🔍 Code Quality Report�[0;35m[MONITOR]�[0m Code Review Monitoring Report �[0;34m[INFO]�[0m Latest Quality Status: �[0;34m[INFO]�[0m Recent monitoring activity: 📈 Current Quality Metrics
Generated on: Sat Jan 24 02:52:53 UTC 2026 Generated by AI DevOps Framework Code Review Monitoring |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request is a valuable documentation update, adding several new browser automation tools to the README and reordering them based on testing. The added rationale for the preference order is very helpful for users to make informed decisions. I've provided one suggestion to improve the readability and scannability of the new list by restructuring the dense descriptions into sub-bullets, which should make the features of each tool clearer.
README.md
Outdated
| - [Agent Browser](https://github.com/vercel-labs/agent-browser) - **DEFAULT** - CLI-first headless automation optimized for AI agents (Rust CLI + Node.js daemon, snapshot+ref pattern for deterministic element targeting, multi-session isolation, zero setup) | ||
| - [Dev-Browser](https://github.com/SawyerHood/dev-browser) - Stateful TypeScript API with persistent Playwright server (14% faster, 39% cheaper than Playwright MCP; pages/cookies/auth survive across script executions; ideal for local dev testing) | ||
| - [Playwriter](https://github.com/remorses/playwriter) - Chrome extension MCP with full Playwright API (reuse existing browser sessions/extensions/cookies, bypass automation detection, single `execute` tool vs 17+ for minimal context bloat) | ||
| - [Sweet Cookie](https://github.com/steipete/sweet-cookie) - Browser cookie extraction for API calls without launching a browser (TypeScript cross-platform + Swift macOS native; reads cookies directly from browser profile DBs) | ||
| - [Stagehand](https://github.com/browserbase/stagehand) - AI-powered natural language browser automation with self-healing selectors (JavaScript + [Python](https://github.com/anthropics/stagehand-python) versions; `act`/`extract`/`observe`/`agent` primitives) | ||
| - [Crawl4AI](https://github.com/unclecode/crawl4ai) - Async web crawler optimized for AI content extraction and bulk scraping | ||
| - [Playwright](https://playwright.dev/) - Cross-browser testing and automation (traditional selector-based approach) | ||
| - [Chrome DevTools](https://chromedevtools.github.io/devtools-protocol/) - Browser debugging, performance analysis, DOM inspection | ||
| - [Cloudflare Browser Rendering](https://developers.cloudflare.com/browser-rendering/) - Server-side web scraping via Cloudflare Workers |
There was a problem hiding this comment.
This is a great update that adds more tools and a helpful preference order. The descriptions are very informative.
To improve scannability and readability, I suggest restructuring the list items with long descriptions to use sub-bullets. This makes the key features of each tool easier to digest at a glance.
I've also capitalized the start of each sub-point for consistency. Here is a suggested refactoring:
- [Agent Browser](https://github.com/vercel-labs/agent-browser) - **DEFAULT** - CLI-first headless automation optimized for AI agents
- Rust CLI + Node.js daemon
- Snapshot+ref pattern for deterministic element targeting
- Multi-session isolation
- Zero setup
- [Dev-Browser](https://github.com/SawyerHood/dev-browser) - Stateful TypeScript API with persistent Playwright server
- 14% faster, 39% cheaper than Playwright MCP
- Pages, cookies, and auth survive across script executions
- Ideal for local dev testing
- [Playwriter](https://github.com/remorses/playwriter) - Chrome extension MCP with full Playwright API
- Reuse existing browser sessions, extensions, and cookies
- Bypasses automation detection
- Single `execute` tool vs 17+ for minimal context bloat
- [Sweet Cookie](https://github.com/steipete/sweet-cookie) - Browser cookie extraction for API calls without launching a browser
- TypeScript cross-platform + Swift macOS native
- Reads cookies directly from browser profile DBs
- [Stagehand](https://github.com/browserbase/stagehand) - AI-powered natural language browser automation with self-healing selectors
- JavaScript + [Python](https://github.com/anthropics/stagehand-python) versions
- `act`/`extract`/`observe`/`agent` primitives
- [Crawl4AI](https://github.com/unclecode/crawl4ai) - Async web crawler optimized for AI content extraction and bulk scraping
- [Playwright](https://playwright.dev/) - Cross-browser testing and automation (traditional selector-based approach)
- [Chrome DevTools](https://chromedevtools.github.io/devtools-protocol/) - Browser debugging, performance analysis, DOM inspection
- [Cloudflare Browser Rendering](https://developers.cloudflare.com/browser-rendering/) - Server-side web scraping via Cloudflare Workers![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Add 6 missing MCP servers to the 'All Supported MCPs' table: - FluentCRM (WordPress CRM) - osgrep (local semantic search) - Peekaboo (macOS GUI automation) - Playwriter (Chrome extension MCP) - WordPress MCP Adapter (official WP MCP) - Move llm-tldr to alphabetical position Also alphabetize the table, add Peekaboo to Browser Automation category, add WordPress MCP Adapter and new CRM & Marketing category section.
🤖 Augment PR SummarySummary: Updates the README’s Browser Automation section to include four additional tools and reorder the full list by tested preference. 🤖 Was this summary useful? React with 👍 or 👎 |
![augmentcode[bot]](https://avatars.githubusercontent.com/in/1027498?s=60&v=4){kind=small}
README.md
Outdated
| - [Cloudflare Browser Rendering](https://developers.cloudflare.com/browser-rendering/) - Server-side web scraping | ||
| **Browser Automation** (in order of preference, tested in OpenCode sessions): | ||
|
|
||
| - [Agent Browser](https://github.com/vercel-labs/agent-browser) - **DEFAULT** - CLI-first headless automation optimized for AI agents (Rust CLI + Node.js daemon, snapshot+ref pattern for deterministic element targeting, multi-session isolation, zero setup) |
There was a problem hiding this comment.
Agent Browser is described as “zero setup”, but the repo’s own guidance indicates a one-time install + Chromium download step (e.g., agent-browser install). Consider clarifying what “zero setup” means here (e.g., no server/config) so readers don’t expect it to run without installation.
🤖 Was this useful? React with 👍 or 👎
🔍 Code Quality Report�[0;35m[MONITOR]�[0m Code Review Monitoring Report �[0;34m[INFO]�[0m Latest Quality Status: �[0;34m[INFO]�[0m Recent monitoring activity: 📈 Current Quality Metrics
Generated on: Sat Jan 24 02:56:21 UTC 2026 Generated by AI DevOps Framework Code Review Monitoring |
- Keep main's tiered MCP table structure (Global vs Per-agent) - Keep main's benchmarked browser automation section with anti-detect stack - Add Peekaboo and Sweet Cookie from PR to browser automation - Add FluentCRM and CRM & Marketing category from PR - Keep WordPress MCP Adapter from both (already in main)
|
🔍 Code Quality Report�[0;35m[MONITOR]�[0m Code Review Monitoring Report �[0;34m[INFO]�[0m Latest Quality Status: �[0;34m[INFO]�[0m Recent monitoring activity: 📈 Current Quality Metrics
Generated on: Sun Jan 25 02:25:23 UTC 2026 Generated by AI DevOps Framework Code Review Monitoring |
1 participant
Summary
Preference Order (from testing)
Why
The README's Browser Automation section was incomplete (missing 4 tools) and didn't reflect the preference order documented in
.agent/tools/browser/browser-automation.mdwhich was established through OpenCode session testing.Changes
README.md- Updated Browser Automation section (13 insertions, 8 deletions)Summary by CodeRabbit
✏️ Tip: You can customize this high-level summary in your review settings.