llms.txt
Compare with other sites' llms.txt files →# DialogueDB
> DialogueDB is a managed database and API for AI conversation management. Store, retrieve, and search multi-turn conversations with threading, state, memory, and semantic search.
## Pages
- [Home](https://dialoguedb.com/): Product overview and features
- [Products](https://dialoguedb.com/products): Platform capabilities and features
- [Integrations](https://dialoguedb.com/integrations): Framework guides for wiring DialogueDB into an existing AI stack
- [Pricing](https://dialoguedb.com/pricing): Plans and pricing
- [Support](https://dialoguedb.com/support): Support resources and contact
- [Blog](https://dialoguedb.com/blog): Product updates and technical articles
## Integrations
- [Vercel AI SDK](https://dialoguedb.com/integrations/vercel-ai-sdk): Every chat survives a Vercel deploy. Users pick up mid-thread on the next visit, and your app can search everything they have discussed before.
- [LangChain](https://dialoguedb.com/integrations/langchain): A drop-in for LangChain's default chat history. Every chain and agent keeps its memory across restarts, and past runs stay searchable by meaning.
- [OpenAI Agents SDK](https://dialoguedb.com/integrations/openai-agents-sdk): The session backend for when your agents outgrow SQLiteSession. Every user stays isolated, every deploy keeps history intact, and past conversations stay searchable.
- [Anthropic SDK](https://dialoguedb.com/integrations/anthropic-sdk): A store for the Claude messages you send. Users are remembered across sessions, past chats stay searchable, and Anthropic's prompt caching keeps full-history replays cheap.
- [Google Gemini SDK](https://dialoguedb.com/integrations/gemini): Persistent chat for apps calling Gemini directly. Every message the SDK sends is stored, users are remembered across sessions, and past chats stay searchable.
- [Grok (xAI)](https://dialoguedb.com/integrations/xai): For chatbots built on Grok. The xAI API is OpenAI-compatible, so the same integration works: chats persist, users are remembered, and past runs stay searchable.
## Blog
- [Introducing DialogueDB](https://dialoguedb.com/blog/introducing-dialoguedb): DialogueDB is live - the conversation infrastructure layer that gives AI applications the ability to remember.
- [How to Store AI Chat History](https://dialoguedb.com/blog/how-to-store-ai-chat-history): Compare storing AI conversation history with PostgreSQL, Redis, DynamoDB, and DialogueDB. Code examples, tradeoffs, and a decision matrix.
- [Building a Stateful AI Chatbot on AWS Lambda](https://dialoguedb.com/blog/stateful-ai-chatbot-aws-lambda): Lambda is stateless. Every invocation starts fresh. Here's how to give your Lambda-based chatbot persistent conversation memory, state, and semantic search.
- [Building an AI Agent That Learns From Its Own Conversations](https://dialoguedb.com/blog/ai-agent-learns-from-conversations): Most conversation storage is passive. The agent has no say. Here's how to give your AI agent autonomous access to its own conversation history.
- [DialogueDB Heads to AWS re:Invent 2025](https://dialoguedb.com/blog/aws-reinvent-2025): Fresh off our launch, the DialogueDB team is heading to AWS re:Invent to meet the developers building the next generation of AI applications.
- [Conversation Persistence in TypeScript AI Agent Frameworks](https://dialoguedb.com/blog/conversation-persistence-typescript-agent-frameworks): TypeScript AI agent frameworks handle sessions well but leave conversation persistence to you, creating gaps in search, threading, and multi-tenancy.
- [DialogueDB Client v2.0.0: CLI, Better Search, Stricter Types](https://dialoguedb.com/blog/client-v2-release): The v2.0.0 release of the DialogueDB TypeScript SDK adds a full CLI, overhauls search with operator-based filtering, and tightens the type system.
- [RAG vs Memory: Why AI Agents Lose Context Over Time](https://dialoguedb.com/blog/retrieval-is-not-memory): RAG retrieves, but it doesn't remember. Here's how teams architect agent memory across extracted facts and conversation databases.
- [Where Should Your Agent's Memory Live?](https://dialoguedb.com/blog/where-should-agent-memory-live): A decision framework for choosing agent memory storage: flat files, SQLite, Postgres, vector databases, and graph databases compared with concrete tradeoffs.
- [Searching Conversation Data with Filters and Natural Language Dates](https://dialoguedb.com/blog/search-api-filters-dates-ordering): Pinpoint conversations by date, tags, and metadata in one DialogueDB Search call, with natural language dates and operators handled server-side instead of in your application code.
- [MCP Is Going Stateless: How to Handle Conversation Persistence](https://dialoguedb.com/blog/mcp-stateless-conversation-persistence): MCP is going stateless at the protocol layer. Here's what that changes for conversation history and how to add persistence without building your own infrastructure.
## Documentation
- [Docs](https://docs.dialoguedb.com): Full documentation site
- [Docs (LLM-friendly)](https://docs.dialoguedb.com/llms-full.txt): Complete documentation in a single markdown file
## Social
- [Twitter](https://x.com/dialoguedbapp)
- [GitHub](https://github.com/dialoguedb)
- [YouTube](https://www.youtube.com/@DialogueDB)
## Blog posts
- [Add Persistent Chat History to the Vercel AI SDK](https://dialoguedb.com/blog/add-persistent-chat-history-vercel-ai-sdk.md): A working example that persists a Vercel AI SDK chat, reloads it after a restart, and adds search and memory over the same messages.
- [Building an AI Agent That Learns From Its Own Conversations](https://dialoguedb.com/blog/ai-agent-learns-from-conversations.md): Most conversation storage is passive. The agent has no say. Here's how to give your AI agent autonomous access to its own conversation history.
- [DialogueDB Heads to AWS re:Invent 2025](https://dialoguedb.com/blog/aws-reinvent-2025.md): Fresh off our launch, the DialogueDB team is heading to AWS re:Invent to meet the developers building the next generation of AI applications.
- [DialogueDB Client v2.0.0: CLI, Better Search, Stricter Types](https://dialoguedb.com/blog/client-v2-release.md): The v2.0.0 release of the DialogueDB TypeScript SDK adds a full CLI, overhauls search with operator-based filtering, and tightens the type system.
- [Conversation Persistence in TypeScript AI Agent Frameworks](https://dialoguedb.com/blog/conversation-persistence-typescript-agent-frameworks.md): TypeScript AI agent frameworks handle sessions well but leave conversation persistence to you, creating gaps in search, threading, and multi-tenancy.
- [Add Durable Session Storage to the OpenAI Agents SDK](https://dialoguedb.com/blog/durable-session-storage-openai-agents-sdk.md): A working example that persists OpenAI Agents SDK runs with tool calls intact, reloads them after a redeploy, and adds memory and semantic search.
- [How to Store AI Chat History: 4 Approaches Compared](https://dialoguedb.com/blog/how-to-store-ai-chat-history.md): Compare storing AI conversation history with PostgreSQL, Redis, DynamoDB, and DialogueDB. Code examples, tradeoffs, and a decision matrix.
- [Introducing DialogueDB: The Database for AI Conversations](https://dialoguedb.com/blog/introducing-dialoguedb.md): DialogueDB is live - the conversation infrastructure layer that gives AI applications the ability to remember.
- [MCP Is Going Stateless: How to Handle Conversation Persistence](https://dialoguedb.com/blog/mcp-stateless-conversation-persistence.md): MCP is going stateless at the protocol layer. Here's what that changes for conversation history and how to add persistence without building it yourself.
- [Multi-Agent Memory Architecture: Scoping Shared State Across AI Agents](https://dialoguedb.com/blog/multi-agent-memory-architecture.md): How to design memory scopes for multi-agent systems: private, shared, and global memory patterns with TypeScript examples.
- [Multi-Tenant Conversation Isolation with Postgres RLS and pgvector](https://dialoguedb.com/blog/postgres-rls-pgvector-multi-tenant-isolation.md): An app-level WHERE clause is one missed filter away from a similarity search returning another tenant's embeddings. RLS fixes that at the database layer.
- [RAG vs Memory: Why AI Agents Lose Context Over Time](https://dialoguedb.com/blog/retrieval-is-not-memory.md): RAG retrieves, but it doesn't remember. Here's how teams architect agent memory across extracted facts and conversation databases.
- [Searching Conversation Data with Filters and Natural Language Dates](https://dialoguedb.com/blog/search-api-filters-dates-ordering.md): Filter conversations by date, tags, and metadata in one DialogueDB Search call — natural language dates and operators handled server-side, not in app code.
- [Building a Stateful AI Chatbot on AWS Lambda](https://dialoguedb.com/blog/stateful-ai-chatbot-aws-lambda.md): Lambda is stateless. Every invocation starts fresh. Here's how to give your Lambda-based chatbot persistent conversation memory, state, and semantic search.
- [Where Should Your Agent's Memory Live?](https://dialoguedb.com/blog/where-should-agent-memory-live.md): A decision framework for choosing agent memory storage: flat files, SQLite, Postgres, vector databases, and graph databases compared with concrete tradeoffs.