Building AI applications and agents
The documentation provides a comprehensive TypeScript toolkit and standardized integration methods, enabling developers to construct AI-powered applications and agents efficiently across various frameworks.
All sources
AI SDK
https://ai-sdk.devAI SDK Documentation for AI Agents & MCP
The AI SDK documentation guides developers in building and integrating AI applications, agents, and custom providers. Learn to implement LLM text generation, streaming, and agent skills, and troubleshoot common issues to build AI-powered applications and agents.
- Pages
- 535
- 4,201
- Tokens
- ~1.1M
- Updated
- Sep 29, 2026
- Monitoring
- Active
Works with
- ChatGPT
- Claude
- Cursor
- MCP clients
Solutions
What AI Agents Can Do With This Source
Use AI SDK Documentation to complete source-specific tasks, make informed decisions, and follow evidence-backed workflows.
Integrating diverse AI models and providers
The AI SDK documentation details how to use the AI Gateway and other providers, offering a unified interface to connect with various AI models without separate integrations.
Generating text with Large Language Models
The `generateText` API reference and cookbook examples provide instructions and parameters for generating text and tool calls from LLMs for non-interactive tasks like drafting or summarizing.
Adding skills and tool calling to AI agents
The documentation on Agent Skills and tool calling provides guides and reference material for extending agents with specialized capabilities and enabling them to interact with external functions.
Optimizing AI application performance with caching
The caching middleware cookbook example demonstrates how to use `LanguageModelMiddleware` to cache assistant responses in KV storage, improving application performance for `generateText` and `streamText` functions.
Troubleshooting slow streaming from Azure OpenAI
The troubleshooting guide offers solutions, such as updating content filtering settings in Azure AI Studio or using the `smoothStream` transformation, to address slow streaming issues with Azure OpenAI.
Coverage
What's Indexed from AI SDK
A quick look at the main areas Bulkgrid has indexed and keeps ready for AI search, MCP, extraction, and change detection.
Docs
/content/docs
291 pages
Introduces the AI SDK as a TypeScript toolkit for building AI applications and agents, covering foundational concepts, core functionalities, UI components, and integration with various providers and harnesses.
Providers
/content/providers
151 pages
Details how to integrate and manage various AI model providers and agent harnesses, including the AI Gateway, community-contributed providers, and observability tools.
Reference
/content/docs/07-reference
135 pages
Provides API reference documentation for core AI SDK functionalities, UI components, server components, workflow tools, and error handling, aiding developers in implementing advanced AI features.
Cookbook
/content/cookbook
93 pages
Offers practical guides and examples for building AI applications, including adding agent skills, implementing caching middleware, calling tools, and setting up API servers across different frameworks.
Reference — AI SDK Core
/content/docs/07-reference/01-ai-sdk-core
62 pages
Documents core AI SDK functions like generateText, streamText, embed, rerank, and generateSpeech, enabling developers to build interactive and automated AI applications.
Community Providers
/content/providers/05-community-providers
60 pages
Guides developers on writing and publishing custom AI SDK providers using the Language Model Specification, fostering community contributions and extending model compatibility.
AI SDK Providers
/content/providers/01-ai-sdk-providers
45 pages
Details the integration of various AI model providers, including AI Gateway, OpenAI, Azure OpenAI, Anthropic, and Amazon Bedrock, simplifying access to diverse LLMs.
AI SDK Errors
/content/docs/07-reference/05-ai-sdk-errors
38 pages
Documents various AI SDK error types, such as AI_APICallError and AI_InvalidArgumentError, providing properties and guidance for checking and resolving common API and data-related issues.
Docs — AI SDK Core
/content/docs/03-ai-sdk-core
32 pages
Provides an overview of AI SDK Core, detailing functions for generating text and structured data, implementing tool calling, and managing runtime and tool context for LLM integration.
Troubleshooting
/content/docs/09-troubleshooting
32 pages
Offers solutions for common AI SDK issues, including slow streaming from Azure OpenAI, problems with server actions in client components, and errors related to stream output and deployment.
Node
/content/cookbook/05-node
30 pages
Provides examples and guides for implementing AI SDK functionalities in Node.js environments, covering text generation, streaming, retrieval augmented generation, and building knowledge base agents.
Next
/content/cookbook/01-next
23 pages
Offers practical examples for integrating the AI SDK with Next.js, demonstrating text and image generation, streaming, caching middleware, and building interactive chatbots.
Advantages
Why Use Bulkgrid for AI-Ready Sources
Bulkgrid turns important websites into fresh, queryable, AI-ready sources without making teams maintain the infrastructure.
Spend Fewer Tokens
Retrieve the right source context instead of sending large pages, pasted docs, or noisy search results into every prompt.
Improve Answer Accuracy
Ground responses in current, source-backed evidence to reduce hallucinations and avoid relying on stale model memory.
Use Fresh Source Data
Keep important websites automatically recrawled and updated so AI is not working from old content.
Query Instantly
Search already-indexed source content immediately instead of crawling pages at question time.
No Infrastructure Maintenance
Bulkgrid handles crawling, rendering, retries, storage, indexing, recrawls, and change detection.
Change Detection
See when pages are added, removed, updated, or fail so teams can react without checking manually.
Give Your Agents Up-to-Date Knowledge from AI SDK
Search and use knowledge from AI SDK in ChatGPT, Claude, Cursor, and any MCP-compatible client.