> ## Documentation Index
> Fetch the complete documentation index at: https://docs.langchain.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Changelog

> Log of updates and improvements to our JavaScript/TypeScript packages

<Callout icon="rss" color="#4F46E5" iconType="regular">
  **Subscribe**: Our changelog includes an [RSS feed](https://docs.langchain.com/oss/javascript/releases/changelog/rss.xml) that can integrate with [Slack](https://slack.com/help/articles/218688467-Add-RSS-feeds-to-Slack), [email](https://zapier.com/apps/email/integrations/rss/1441/send-new-rss-feed-entries-via-email), Discord bots like [Readybot](https://readybot.io/) or [RSS Feeds to Discord Bot](https://rss.app/en/bots/rssfeeds-discord-bot), and other subscription tools.
</Callout>

<Update label="Mar 24, 2026" tags={["deepagents"]} rss={{ title: "Mar 24, 2026 - deepagents" }}>
  ## `deepagents` v1.9.0-alpha.0

  Alpha release of `deepagents` v1.9.0.

  * **[Async subagents](/oss/javascript/deepagents/async-subagents)**: Deep Agents can launch non-blocking background tasks, so users can continue interacting with the agent while subagents work concurrently. Requires [LangSmith Deployment](/langsmith/deployment) for sub-agents.

  * **[Backend](/oss/javascript/deepagents/backends) protocol v2**: We've introduced a new v2 backend protocol (`BackendProtocolV2`) with backward-compatible changes to the Deep Agents backend interface. Key changes:
    * **Structured result types**: All methods now return structured `Result` objects (e.g., `ReadResult`, `LsResult`, `GrepResult`, `GlobResult`) with consistent error handling via an `error` field, instead of returning raw values or throwing exceptions.
    * **Multi-modal file support**: `read()` returns a `ReadResult` with a `.content` field instead of a plain string. For binary files (images, PDFs, audio, video), the full raw `Uint8Array` content is returned via `readRaw()`, enabling agents to work with multi-modal files natively.
    * **Simplified method names**: `lsInfo` -> `ls`, `grepRaw` -> `grep`, `globInfo` -> `glob`.
    * **Backward compatible**: Existing v1 backends can be adapted to the v2 interface using `adaptBackendProtocol`. The v1 interfaces (`BackendProtocolV1`, `SandboxBackendProtocolV1`) are deprecated but retained for compatibility.
</Update>

<Update label="Jan 14, 2026" tags={["langgraph"]} rss={{ title: "Jan 14, 2026 - langgraph" }}>
  ## v1.1.0

  ### `@langchain/langgraph`

  Introducing **StateSchema** - a cleaner, library-agnostic way to define graph state that works with any [Standard Schema](https://github.com/standard-schema/standard-schema)-compliant validation library.

  ### Standard JSON Schema support

  LangGraph now supports [Standard JSON Schema](https://standardschema.dev/json-schema), an open specification implemented by Zod 4, Valibot, ArkType, and other schema libraries. This means you can use your preferred validation library without lock-in:

  ```typescript theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
  import { z } from "zod"; // or valibot, arktype, etc.
  import { StateSchema, ReducedValue, MessagesValue } from "@langchain/langgraph";

  const AgentState = new StateSchema({
    messages: MessagesValue,
    currentStep: z.string(),
    count: z.number().default(0),
    history: new ReducedValue(
      z.array(z.string()).default(() => []),
      {
        inputSchema: z.string(),
        reducer: (current, next) => [...current, next],
      }
    ),
  });

  // Type-safe state and update types
  type State = typeof AgentState.State;
  type Update = typeof AgentState.Update;

  const graph = new StateGraph(AgentState)
    .addNode("agent", (state) => ({ count: state.count + 1 }))
    .addEdge(START, "agent")
    .addEdge("agent", END)
    .compile();
  ```

  ### New state value primitives

  * **ReducedValue**: Define fields with custom reducers for accumulating values. Supports separate input and output schemas for type-safe reducer inputs.
  * **UntrackedValue**: Define transient state that exists during execution but is never checkpointed - useful for database connections, caches, or runtime-only configuration.
  * **MessagesValue**: A prebuilt `ReducedValue` for chat messages with the standard messages reducer.

  ### Type helper exports

  New exported type utilities for typing functions outside the graph builder:

  * `GraphNode<Schema, Nodes?, Config?>` - Type node functions with full inference
  * `ConditionalEdgeRouter<Schema, Nodes?>` - Type conditional edge routers

  ```typescript theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
  // Type standalone node functions
  const myNode: GraphNode<typeof AgentState> = (state, config) => {
    return { count: state.count + 1 };
  };

  // Use schema type helpers directly
  const processState = (state: typeof AgentState.State) => {
    console.log(state.count);
  };
  ```

  The existing `Annotation` and zod-based API continues to work unchanged - `StateSchema` is an additional option for those who prefer schema-first definitions.

  <Card title="Learn more about StateSchema" icon="book" href="/oss/javascript/langgraph/graph-api#schema">
    See the full documentation for defining graph state with StateSchema, ReducedValue, and UntrackedValue.
  </Card>

  <Card title="Learn about type utilities" icon="code" href="/oss/javascript/langgraph/graph-api#type-utilities">
    Use GraphNode and ConditionalEdgeRouter to type functions outside the graph builder.
  </Card>
</Update>

<Update label="Dec 12, 2025" tags={["langchain", "@langchain/openai", "@langchain/anthropic", "@langchain/ollama", "@langchain/community", "@langchain/xai", "@langchain/tavily", "@langchain/mongodb", "@langchain/mcp-adapters", "@langchain/google-common", "@langchain/core"]} rss={{ title: "Dec 12, 2025 - langchain" }}>
  ## v1.2.0

  ### `langchain`

  * [Structured output](/oss/javascript/langchain/structured-output): Added ability to manually set `strict` mode when using `providerStrategy` for structured output.

  ### `@langchain/openai`

  * **New provider built-in tools:** Support for file search, web search, code interpreter, image generation, computer use, shell, and MCP connector tools executed server-side by the provider. See [Server-side tool use](/oss/javascript/langchain/tools#server-side-tool-use) and the [OpenAI](/oss/javascript/integrations/chat/openai) chat integration.
  * **Content moderation:** New `moderateContent` option on `ChatOpenAI` for detecting and handling unsafe content.
  * Prefer responses API for GPT-5.2 Pro model.

  ## v1.3.0

  ### `@langchain/anthropic`

  * **New provider built-in tools:** Support for text editor, web fetch, computer use, tool search, and MCP toolset tools executed server-side by the provider. See [Server-side tool use](/oss/javascript/langchain/tools#server-side-tool-use) and the [Anthropic](/oss/javascript/integrations/chat/anthropic) chat integration.
  * Exposed `ChatAnthropicInput` type for improved type safety.

  ## v1.1.0

  ### `@langchain/ollama`

  * **Native structured outputs:** Added support for native structured output via `withStructuredOutput`.
  * Support for custom `baseUrl` configuration.

  ## v1.0.0

  ### `@langchain/community`

  * Jira document loader updated to use v3 API.
  * LanceDB: Added `similaritySearch()` and `similaritySearchWithScore()` support.
  * Elasticsearch hybrid search support.
  * New `GoogleCalendarDeleteTool`.
  * Various bug fixes for LlamaCppEmbeddings, PrismaVectorStore, IBM WatsonX, and security improvements.

  ### Other packages

  * **@langchain/xai:** Native Live Search support.
  * **@langchain/tavily:** Added Tavily's research endpoint.
  * **@langchain/mongodb:** New MongoDB LLM cache.
  * **@langchain/mcp-adapters:** Added `onConnectionError` option.
  * **@langchain/google-common:** `jsonSchema` method support in `withStructuredOutput`.
  * **@langchain/core:** Security fixes, better subgraph nesting in Mermaid graphs, UUID7 for run IDs.
</Update>

<Update label="Nov 25, 2025" tags={["langchain"]} rss={{ title: "Nov 25, 2025 - langchain" }}>
  ## v1.1.0

  * [Model profiles](/oss/javascript/langchain/models#model-profiles): Chat models now expose supported features and capabilities through a `.profile` getter. These data are derived from [models.dev](https://models.dev), an open source project providing model capability data.
  * [Model retry middleware](/oss/javascript/langchain/middleware/built-in#model-retry): New middleware for automatically retrying failed model calls with configurable exponential backoff, improving agent reliability.
  * [Content moderation middleware](/oss/javascript/langchain/middleware/built-in#provider-specific-middleware): OpenAI content moderation middleware for detecting and handling unsafe content in agent interactions. Supports checking user input, model output, and tool results.
  * [Summarization middleware](/oss/javascript/langchain/middleware/built-in#summarization): Updated to support flexible trigger points using model profiles for context-aware summarization.
  * [Structured output](/oss/javascript/langchain/structured-output): `ProviderStrategy` support (native structured output) can now be inferred from model profiles.
  * [`SystemMessage` for `createAgent`](/oss/javascript/langchain/middleware/custom#dynamic-prompt): Support for passing `SystemMessage` instances directly to `createAgent`'s `systemPrompt` parameter and a new `concat` method for extending system messages. Enables advanced features like cache control and structured content blocks.
  * [Dynamic system prompt middleware](/oss/javascript/langchain/agents#dynamic-system-prompt): Return values from `dynamicSystemPromptMiddleware` are now purely additive. When returning a [`SystemMessage`](https://reference.langchain.com/javascript/langchain-core/messages/SystemMessage) or `string`, they are merged with existing system messages rather than replacing them, making it easier to compose multiple middleware that modify the prompt.
  * **Compatibility improvements:** Fixed error handling for Zod v4 validation errors in structured output and tool schemas, ensuring detailed error messages are properly displayed.
</Update>

<Update label="Oct 20, 2025" tags={["langchain", "langgraph"]} rss={{ title: "Oct 20, 2025 - langchain" }}>
  ## v1.0.0

  ### `langchain`

  * [Release notes](/oss/javascript/releases/langchain-v1)
  * [Migration guide](/oss/javascript/migrate/langchain-v1)

  ### `langgraph`

  * [Release notes](/oss/javascript/releases/langgraph-v1)
  * [Migration guide](/oss/javascript/migrate/langgraph-v1)

  <Callout icon="speakerphone" color="#4F46E5" iconType="regular">
    If you encounter any issues or have feedback, please [open an issue](https://github.com/langchain-ai/docs/issues/new?template=01-langchain.yml) so we can improve. To view v0.x documentation, [go to the archived content](https://github.com/langchain-ai/langchainjs/tree/v0.3/docs/core_docs/docs).
  </Callout>
</Update>

***

<div className="source-links">
  <Callout icon="terminal-2">
    [Connect these docs](/use-these-docs) to Claude, VSCode, and more via MCP for real-time answers.
  </Callout>

  <Callout icon="edit">
    [Edit this page on GitHub](https://github.com/langchain-ai/docs/edit/main/src/oss/javascript/releases/changelog.mdx) or [file an issue](https://github.com/langchain-ai/docs/issues/new/choose).
  </Callout>
</div>