added braintrust and agentkit readmes (#1)

Author: Kylejeong2

README.md

@@ -63,7 +63,7 @@ Enhance your Vercel applications with web-browsing capabilities. Build Generativ
 
 ### 📊 Evaluation & Testing
 
-#### [**Braintrust Integration**](./examples/integrations/braintrust/)
+#### [**Braintrust Integration**](./examples/integrations/braintrust/README.md)
 Integrate Browserbase with Braintrust for evaluation and testing of AI agent performance in web environments. Monitor, measure, and improve your browser automation workflows.
 
 ## 🏗️ Monorepo Structure

examples/integrations/agentkit/README.md

@@ -0,0 +1,39 @@
+# AgentKit
+
+## Using AgentKit with Browserbase
+
+Develop AI Agents that can browse the web autonomously using Browserbase and AgentKit (by inngest).
+
+## Overview
+
+By the end of this guide, you'll have an AI agent built with AgentKit that can browse the web using Browserbase.
+
+### You'll learn how to:
+
+- Create AgentKit tools that leverage Browserbase's managed headless browsers
+- Build autonomous web browsing agents that can search, extract data, and interact with websites 
+- Use Stagehand, Browserbase's AI-powered browser automation library, to create resilient web agents
+
+### This integration is useful for:
+
+- Building AI agents that can fetch real-time information from the web
+- Creating autonomous browsing capabilities within your agents
+- Developing resilient web scraping and interaction systems
+- Enabling your agents to perform complex web-based tasks
+
+### Integration Approaches
+
+There are two approaches to using Browserbase with AgentKit:
+
+1. **Create custom Browserbase tools**: Ideal for simple actions on webpages with manual browser control
+2. **Use Stagehand library as tools**: Better for autonomous browsing and resilient web scraping
+
+### Prerequisites
+
+Before you start, make sure you have:
+
+- AgentKit installed
+- Browserbase Project ID & API key  
+- (Optional) LLM API key of your choice to use with Stagehand
+
+Next, let's dive into building web browsing agents with AgentKit and Browserbase.

examples/integrations/braintrust/README.md

@@ -0,0 +1,30 @@
+# Braintrust Integration
+
+Utilize Browserbase with Braintrust for advanced browser automation.
+
+## Overview
+
+Braintrust is a platform for building AI applications, making it more efficient for Large Language Models (LLMs) to become more robust, reliable, and interactive.
+
+### Key Features
+
+- Prototyping with different prompts and LLMs in a sandboxed environment
+- Real-time monitoring and performance insights  
+- Data management through intuitive UI
+
+## Tool Calling Support
+
+Most Large Language Models support tool calling, which allows you to define tools with well-defined input and output types. Common use cases include:
+
+1. Enabling models to "call" tools that perform external tasks and use those results to produce a final response
+2. Coercing models into producing structured outputs that match a given JSON schema
+
+Braintrust supports both use cases, allowing prompts to be executed directly using TypeScript or Python.
+
+### Custom Functions
+
+Custom functions enhance LLM capabilities by enabling:
+
+- Web-browsing capabilities (using tools like Browserbase)
+- Complex computations
+- External API access