diff --git a/agents/browser-use__browser-use-web-ui/README.md b/agents/browser-use__browser-use-web-ui/README.md
new file mode 100644
index 0000000..7e2eb75
--- /dev/null
+++ b/agents/browser-use__browser-use-web-ui/README.md
@@ -0,0 +1,70 @@
+# Browser Use Web UI
+
+**Repo:** https://github.com/browser-use/web-ui  ·  ⭐ 16 000+  ·  MIT
+
+A **Gradio web interface** for [browser-use](https://github.com/browser-use/browser-use)
+that lets any LLM (OpenAI, Anthropic, Google Gemini, DeepSeek, Ollama, Mistral,
+Azure OpenAI, and more) autonomously control a real Chromium browser.
+
+---
+
+## What it does
+
+Give it a task in plain English. It figures out the browser actions needed —
+clicking, typing, navigating, scrolling, extracting data — and executes them
+step-by-step inside a live browser window you can watch in real time.
+
+### Two agent modes
+
+| Mode | Description |
+|---|---|
+| **Browser Agent** | Single-goal autonomous browsing. Give it a task; it completes it. |
+| **Deep Research Agent** | Multi-tab, multi-step research orchestrated by LangGraph. Produces a cited Markdown report. |
+
+---
+
+## Key capabilities
+
+- 🌐 **Multi-LLM** — plug in any LangChain-compatible model via `.env` or the Settings tab
+- 🖥️ **Live browser view** — watch the agent work, pause/resume mid-task
+- 🔐 **Own-browser support** — connect your existing Chrome profile so the agent inherits your logins
+- 📹 **GIF recordings** — every session is recorded for audit/replay
+- 🔌 **MCP tool support** — invoke Model Context Protocol tools mid-task
+- 📝 **Persistent sessions** — keep the browser open between tasks (no re-login)
+- 🐳 **Docker-ready** — ships a `Dockerfile` and `docker-compose.yml` for one-command deployment
+
+---
+
+## Quick start
+
+```bash
+git clone https://github.com/browser-use/web-ui.git && cd web-ui
+cp .env.example .env          # add your LLM API key
+uv pip install -r requirements.txt
+playwright install chromium --with-deps
+python webui.py --ip 127.0.0.1 --port 7788
+# open http://127.0.0.1:7788
+```
+
+Or with Docker:
+
+```bash
+docker-compose up
+```
+
+---
+
+## Example tasks
+
+- *"Go to amazon.com and find the cheapest USB-C hub with at least 4 ports"*
+- *"Research the top 5 open-source LLM frameworks released in 2024 and write a comparison report"*
+- *"Fill in the contact form at example.com with my details and click Submit"*
+- *"Scrape the pricing table from competitor.io and return it as JSON"*
+
+---
+
+## GitAgent Protocol
+
+This agent ships `agent.yaml` + `SOUL.md` at the repo root, making it
+GAP-conformant and runnable on any compatible runtime.  
+Learn more at [gitagent.sh](https://gitagent.sh).
diff --git a/agents/browser-use__browser-use-web-ui/metadata.json b/agents/browser-use__browser-use-web-ui/metadata.json
new file mode 100644
index 0000000..5059093
--- /dev/null
+++ b/agents/browser-use__browser-use-web-ui/metadata.json
@@ -0,0 +1,15 @@
+{
+  "name": "browser-use-web-ui",
+  "author": "browser-use",
+  "description": "Gradio UI for browser-use: control a real browser with any LLM. Autonomous browsing, form-filling, data extraction, and deep multi-tab research.",
+  "repository": "https://github.com/shreyas-lyzr/web-ui",
+  "path": "",
+  "version": "1.0.0",
+  "category": "developer-tools",
+  "tags": ["browser-automation", "web-scraping", "gradio", "langchain", "deep-research", "multi-llm", "playwright", "form-filling", "data-extraction", "autonomous-agent"],
+  "license": "MIT",
+  "model": "openai:gpt-4o",
+  "adapters": ["system-prompt", "claude-code"],
+  "icon": false,
+  "banner": false
+}