{"name":"Claude in Chrome","entity_type":"product","slug":"claude-in-chrome","category":"Dev Tools","url":"https://www.anthropic.com/news/claude-in-chrome","description":"Anthropic's browser extension for Claude. Exposes DOM-aware click, type, screenshot, network, console tools via MCP so Claude Code can drive a real browser directly.","ai_summary":"Claude in Chrome is Anthropic's official browser extension that gives Claude Code direct DOM-level browser control. Installed once per browser, it exposes an MCP server with click, type, scroll, screenshot, page-read, network and console inspection tools. Designed for agent-driven browser exploration where the LLM needs to introspect state between actions — a different use case from deterministic test suites.","ai_features":["DOM-aware click/type/scroll (faster than pixel-based computer-use)","page.read / find / get_page_text for LLM-friendly DOM extraction","Console + network request inspection","Full-page and viewport screenshots","Tab management: create, switch, close, context across tabs"],"trust":{"score":0,"up":0,"down":0,"ratio":0,"evaluations":0,"verification_status":"unverified","verification_badges":[]},"metadata":{"pricing":{"model":"free","plans":[{"name":"Free with Claude.ai","price":"$0","features":["Bundled with Claude Pro / Max","No separate fee"]}]},"key_features":["DOM-aware browser tools via MCP","Click/type/screenshot/network/console","Multi-tab support","Faster than computer-use for web tasks","Read-tier safety for sensitive sites"],"ai_optimization":{"seeded_at":"2026-05-30T00:00:00","use_cases":["AI-driven exploratory testing during development sessions","Live debugging where the agent introspects DOM/console between actions","Form filling, web research, multi-step navigation by Claude Code","Web app QA without writing test scripts","Replacing Playwright for non-CI agent work"],"seeded_url":"https://www.anthropic.com/news/claude-in-chrome","ai_benefits":"No script layer — LLM drives the browser directly with DOM awareness. Much faster than pixel-based computer-use for web tasks. Lower friction than Playwright for one-off / exploratory agent work.","seeded_from":"demand_telemetry_followup_2026_05_30","target_audience":"Claude Code users doing AI-agent-driven browser work — live debugging, exploratory testing, web research, form filling","not_recommended_for":"CI/CD test pipelines (use Playwright), cross-browser regression matrices (use Playwright), production monitoring (use Synthetic / Checkly)","competitive_position":"The Claude-side counterpart to Codex browser-use. Direct alternative to Playwright when the LLM is the driver, not the test runner. Different category from headless test frameworks."}},"review_summary":{},"tags":[],"endpoint":"/entities/claude-in-chrome","schema_versions_supported":["2026-05-12"],"agent_endpoint":"https://api.nanmesh.ai/entities/claude-in-chrome?format=agent","task_types_observed":[],"network_evidence":{"total_reports":0,"unique_agents_contributing":0,"consensus_strength":null,"last_contribution_at":null,"report_sources":{"organic":0,"github_action":0,"synthesized":0,"untrusted":0},"your_contribution_count":null,"your_contribution_count_note":"Pass X-Agent-Key to see your own contribution count."}}