Cloudflare changelogs | Agents

Agents, Workers - Agent traces for Think, Flue, and AI SDK instrumented by Agents SDK

Tue, 04 Aug 2026 00:00:00 GMT

Agent tracing is now available for applications built with the Agents SDK. Traces show each agent turn alongside model calls, tool runs, approvals, token usage, and Workers runtime operations.

Turn on Workers tracing in your Wrangler configuration:

{
  "$schema": "./node_modules/wrangler/config-schema.json",
  "observability": {
    "traces": {
      "enabled": true
    }
  }
}

[observability.traces]
enabled = true

Think and Flue applications emit agent traces automatically. For direct AI SDK calls, wrap the AI SDK namespace once. wrapAISDK() supports AI SDK v6 and v7. This AI SDK v7 example also supplies the agent identity:

import * as ai from "ai";
import { wrapAISDK } from "agents/observability/ai";

const tracedAI = wrapAISDK(ai);

await tracedAI.generateText({
	model,
	prompt: "Find an available appointment",
	runtimeContext: {
		agentId: "booking-agent-production",
		conversationId: "conversation-123",
	},
	telemetry: {
		functionId: "booking-agent",
		includeRuntimeContext: {
			agentId: true,
			conversationId: true,
		},
	},
});

import * as ai from "ai";
import { wrapAISDK } from "agents/observability/ai";

const tracedAI = wrapAISDK(ai);

await tracedAI.generateText({
	model,
	prompt: "Find an available appointment",
	runtimeContext: {
		agentId: "booking-agent-production",
		conversationId: "conversation-123",
	},
	telemetry: {
		functionId: "booking-agent",
		includeRuntimeContext: {
			agentId: true,
			conversationId: true,
		},
	},
});

Message and tool payload recording is off by default. Turn it on only when the payloads are safe to store:

const tracedAI = wrapAISDK(ai, {
	storeMessages: true,
	storeTools: true,
});

const tracedAI = wrapAISDK(ai, {
	storeMessages: true,
	storeTools: true,
});

Open the Agents tab ↗ in the Cloudflare dashboard to inspect sessions, replay conversations, and view trace waterfalls. For advanced setup, privacy controls, and trace structure, refer to Agent tracing.

Agents, Workers - Preview: @cloudflare/computer agent runtime

Mon, 03 Aug 2026 00:00:00 GMT

We're releasing an early preview of @cloudflare/computer ↗, an open-source agent runtime that gives every agent its own computer. The runtime dynamically orchestrates between fast, efficient isolates and full Linux containers, so the agent always runs on the right compute primitive for the task at hand.

@cloudflare/computer provides a virtual filesystem backed by SQLite, which you can populate from cloud storage, source control, or any files you choose. Agents can read, write, and edit files, run shell commands, and interact with Git repositories. All operations are gated, audited, and observed.

Install the package via npm:

npm install @cloudflare/computer

Instantiate a Workspace inside any Durable Object to give your agent a filesystem and execution runtime:

import { Workspace } from "@cloudflare/computer";

export class Agent {
	workspace = new Workspace({
		storage: this.ctx.storage,
	});
}

Several execution backends are included or you can write your own:

Isolate runtime — fast, horizontally scalable execution via just-bash and Dynamic Workers, ideal for file manipulation and data processing.
Container runtime — full Linux environment via Cloudflare Containers, mounted through FUSE, for tasks that need native binaries, package managers, or a complete userland.

The AI SDK-compatible toolkit provides common agent tools (read, write, edit, ls, exec) and guides the model to choose the appropriate backend for each task.

For more examples, including a step-by-step tutorial, visit the @cloudflare/computer repository ↗.

Read the announcement blog post for more details: Your agent needs a computer, not a container ↗.

Agents, Workers - Cloudflare MCP servers support the new MCP 2026-07-28 Specification

Tue, 28 Jul 2026 00:00:00 GMT

Cloudflare's product-specific MCP servers now support the new MCP 2026-07-28 Specification. Each request runs on a fresh stateless server without an MCP protocol session or protocol-specific Durable Object.

The /mcp endpoint also accepts stateless requests from 2025 Streamable HTTP clients. Most clients can reconnect without configuration changes.

Use /mcp for new connections. Historical /sse URLs continue to work as aliases for the same Streamable HTTP handler, but they no longer serve the deprecated HTTP+SSE transport. If a client forces SSE transport, change it to Streamable HTTP or automatic transport detection.

Agents, Workers - Agents SDK adds MCP Specification 2026-07-28 support

Mon, 27 Jul 2026 00:00:00 GMT

Agents SDK v0.20.0 adds client and server support for the MCP 2026-07-28 release candidate ↗. Workers can serve tools, prompts, resources, and elicitation without an MCP transport session or Durable Object. Agents can connect to both MCP 2026-07-28 servers and existing legacy servers.

Client support

The MCP client manager now uses @modelcontextprotocol/client. For each connection, it probes for MCP 2026-07-28 support with server/discover. If the server does not support the stateless protocol, the client continues with the legacy initialize handshake on the same connection. Existing addMcpServer calls do not need a protocol-version setting or separate clients for each protocol generation.

For stateless requests, elicitation uses input_required through multi-round-trip requests (MRTR). The legacy path uses the same form and URL handlers for pushed requests. The SDK collects input, retries the original operation, and resolves the original callTool, getPrompt, or readResource promise with its final result.

OAuth callbacks now validate issuer metadata through the v2 SDK. Discovery state and issuer-bound credentials persist across browser redirects and Durable Object hibernation.

Run stateless servers

createMcpHandler now accepts a factory that returns a server from @modelcontextprotocol/server. The factory creates an isolated server for each request.

import { McpServer } from "@modelcontextprotocol/server";
import { createMcpHandler } from "agents/mcp/server";

function createServer() {
	return new McpServer({ name: "example", version: "1.0.0" });
}

export default {
	fetch(request, env, ctx) {
		return createMcpHandler(createServer)(request, env, ctx);
	},
};

import { McpServer } from "@modelcontextprotocol/server";
import { createMcpHandler } from "agents/mcp/server";

function createServer() {
	return new McpServer({ name: "example", version: "1.0.0" });
}

export default {
	fetch(request, env, ctx) {
		return createMcpHandler(createServer)(request, env, ctx);
	},
} satisfies ExportedHandler;

The isolated agents/mcp/server entry keeps McpAgent, WorkerTransport, MCP client transports, and SDK v1 modules out of stateless server bundles.

The Workers wrapper validates present browser Origins, supports explicit delegation to trusted Origin middleware, and exposes request handling plus typed change notifications.

Backward compatibility

The same createMcpHandler(createServer)(request, env, ctx) route serves MCP 2026-07-28 clients and legacy clients that use stateless requests. You do not need separate routes or tool definitions for ordinary tools, prompts, and resources.

McpAgent is deprecated and feature-frozen. Migrate existing McpAgent servers to the stateless handler at your earliest convenience. If a server depends on protocol sessions, RPC, pushed server-to-client requests, standalone streams, or replay, use the migration guide to design stateless equivalents and run both routes while clients transition.

Migrate existing SDK v1 servers

Upgrade the Agents SDK:

npm i agents@latest

yarn add agents@latest

pnpm add agents@latest

bun add agents@latest

Move ordinary SDK v1 server definitions into an SDK v2 factory and serve them with createMcpHandler. The handler's default legacy compatibility means most stateless deployments need only one route.

If an existing McpAgent server still needs sessionful features, add the stateless path beside it. Use isLegacyRequest() to send only legacy traffic to the existing route:

import { isLegacyRequest } from "@modelcontextprotocol/server";
import { createMcpHandler } from "agents/mcp/server";
import { MyMcpAgent } from "./legacy-server";
import { createServer } from "./server";

const stateless = createMcpHandler(createServer, {
	route: "/mcp",
	legacy: "reject",
});
const legacy = MyMcpAgent.serve("/mcp");

export default {
	async fetch(request, env, ctx) {
		if (await isLegacyRequest(request)) {
			return legacy.fetch(request, env, ctx);
		}
		return stateless(request, env, ctx);
	},
};

import { isLegacyRequest } from "@modelcontextprotocol/server";
import { createMcpHandler } from "agents/mcp/server";
import { MyMcpAgent } from "./legacy-server";
import { createServer } from "./server";

const stateless = createMcpHandler(createServer, {
	route: "/mcp",
	legacy: "reject",
});
const legacy = MyMcpAgent.serve("/mcp");

export default {
	async fetch(request: Request, env: Env, ctx: ExecutionContext) {
		if (await isLegacyRequest(request)) {
			return legacy.fetch(request, env, ctx);
		}
		return stateless(request, env, ctx);
	},
} satisfies ExportedHandler<Env>;

Migrate the remaining sessionful features, allow existing sessions to drain, then remove the legacy route. Refer to Migrate to MCP SDK v2 for package changes, compatibility limits, and rollout steps.

Deprecations in v0.20.0

This release deprecates the following Agents SDK APIs:

Deprecated API	Replacement	Status
`McpAgent`	Use an SDK v2 factory with `createMcpHandler` for stateless servers. Use the migration guide to replace stateful features before removing a legacy route.	Feature-frozen. No removal version is announced.
`createMcpHandler(v1Server, options)`	Move the server to an SDK v2 factory and call `createMcpHandler(factory, options)`. Use `createLegacyMcpHandler` only as a temporary bridge for sessionful features.	Scheduled for removal in the next major version.
`MCPClientManager.callTool(params, resultSchema, options)` and the equivalent `withX402Client` overload	Use `callTool(params, options)` or `callTool(confirm, params, options)`.	Compatibility overload. No removal version is announced.

The MCP 2026-07-28 draft separately deprecates Roots, Sampling, Logging, the old HTTP+SSE transport, and Dynamic Client Registration.

Agents - Agents SDK packages support AI SDK v6 and v7

Thu, 23 Jul 2026 00:00:00 GMT

The agents, @cloudflare/ai-chat, @cloudflare/codemode, and @cloudflare/think packages now support AI SDK v6 and v7. Existing applications can remain on v6 when updating these packages. Applications can also adopt v7 without changing the Cloudflare Agents APIs they use.

The supported peer ranges are ai@^6 || ^7 and @ai-sdk/react@^3 || ^4. Use matching major versions: pair AI SDK v6 with @ai-sdk/react v3, or pair AI SDK v7 with @ai-sdk/react v4.

To install the latest packages with AI SDK v7:

npm i agents@latest @cloudflare/ai-chat@latest @cloudflare/codemode@latest @cloudflare/think@latest ai@^7 @ai-sdk/react@^4

yarn add agents@latest @cloudflare/ai-chat@latest @cloudflare/codemode@latest @cloudflare/think@latest ai@^7 @ai-sdk/react@^4

pnpm add agents@latest @cloudflare/ai-chat@latest @cloudflare/codemode@latest @cloudflare/think@latest ai@^7 @ai-sdk/react@^4

bun add agents@latest @cloudflare/ai-chat@latest @cloudflare/codemode@latest @cloudflare/think@latest ai@^7 @ai-sdk/react@^4

Think normalizes streaming, tool completion events, and telemetry across both AI SDK versions. Existing v6 applications do not need to migrate these integrations before updating Think.

For setup and usage details, refer to the Think documentation.

Agents, Workers - Agents SDK reduces MCP schema conversion, adds exposure controls for MCP in Think and Code Mode SDK adds direct host APIs

Wed, 22 Jul 2026 00:00:00 GMT

This release reduces repeated MCP schema conversion and adds an opt-out for Think's automatic MCP tool exposure. It also lets non-AI-SDK hosts invoke the durable Code Mode runtime directly.

Control direct MCP tool exposure in Think

Agents SDK MCP clients now reuse converted input and output schemas while a live connection keeps the same tool catalog. This avoids converting every MCP JSON Schema to Zod again for each model turn.

@cloudflare/think also adds includeMcpTools. Set it to false when you expose MCP tools through Code Mode or another mechanism outside Think's automatic tool set:

import { Think } from "@cloudflare/think";

export class MyAgent extends Think {
	includeMcpTools = false;
	waitForMcpConnections = true;
}

import { Think } from "@cloudflare/think";

export class MyAgent extends Think<Env> {
	includeMcpTools = false;
	waitForMcpConnections = true;
}

This setting skips Think's automatic getAITools() call. MCP registration, restoration, discovery, raw catalog access, direct calls, and Code Mode connectors continue to work.

Use listTools() when you only need the raw MCP catalog. For connector setup, refer to Use MCP tools with Code Mode.

Invoke the Code Mode runtime without the AI SDK

@cloudflare/codemode@latest adds execute(), search(), and describe() to the durable runtime handle. MCP servers and other hosts can now execute code and discover connector methods without adapting the runtime to an AI SDK tool.

const matches = await runtime.search("create issue");
const docs = await runtime.describe(matches.results[0].path);
const outcome = await runtime.execute({
	code: `async () => github.create_issue({ title: "Bug" })`,
});

const matches = await runtime.search("create issue");
const docs = await runtime.describe(matches.results[0].path);
const outcome = await runtime.execute({
	code: `async () => github.create_issue({ title: "Bug" })`,
});

Search and describe results include requiresApproval: true for protected connector methods. Resolve a paused execution with the existing approve() and reject() methods.

For setup and exact method types, refer to Create a durable Code Mode runtime and the Code Mode API reference.

Upgrade

npm i agents@latest @cloudflare/think@latest @cloudflare/codemode@latest

yarn add agents@latest @cloudflare/think@latest @cloudflare/codemode@latest

pnpm add agents@latest @cloudflare/think@latest @cloudflare/codemode@latest

bun add agents@latest @cloudflare/think@latest @cloudflare/codemode@latest

Agents, Workers - Agents can respond to MCP elicitation requests

Mon, 13 Jul 2026 00:00:00 GMT

Agents connected to Model Context Protocol (MCP) servers with addMcpServer can now handle elicitation ↗ requests.

Elicitation lets an MCP server request user input while it handles a tool call. Form mode collects structured, non-sensitive data. URL mode asks for consent before opening an out-of-band flow, such as third-party authorization or payment.

sequenceDiagram
    participant User
    participant Agent as Agent (MCP client)
    participant Server as MCP server
    participant Browser

    Server->>Agent: elicitation/create
    Agent->>User: Show server, reason, and input or URL
    User->>Agent: Submit, open, decline, or cancel
    Agent->>Browser: Open URL after consent (URL mode)
    Agent->>Server: accept, decline, or cancel
    Server-->>Agent: Optional URL completion notification

import { Agent } from "agents";

export class MyAgent extends Agent {
	onStart() {
		this.mcp.configureElicitationHandlers({
			form: (request, serverId) => this.forwardToUser(request, serverId),
			url: (request, serverId) => this.forwardToUser(request, serverId),
		});
	}

	forwardToUser(request, serverId) {
		// Show the request in your UI and resolve after the user responds.
		throw new Error(
			`Implement elicitation for ${serverId}: ${request.params.message}`,
		);
	}
}

import { Agent } from "agents";
import type { ElicitRequest, ElicitResult } from "agents/mcp";

export class MyAgent extends Agent<Env> {
	onStart() {
		this.mcp.configureElicitationHandlers({
			form: (request, serverId) => this.forwardToUser(request, serverId),
			url: (request, serverId) => this.forwardToUser(request, serverId),
		});
	}

	private forwardToUser(
		request: ElicitRequest,
		serverId: string,
	): Promise<ElicitResult> {
		// Show the request in your UI and resolve after the user responds.
		throw new Error(
			`Implement elicitation for ${serverId}: ${request.params.message}`,
		);
	}
}

Connections advertise only the modes with configured handlers. An Agent without handlers advertises no elicitation capability, which lets the server use its fallback. The SDK stores the advertised modes with each MCP server registration so they survive Durable Object hibernation. Callback functions remain in memory and reattach when onStart() runs.

For implementation details and a browser forwarding pattern, refer to MCP client elicitation. The mcp-client ↗ and mcp-elicitation ↗ examples implement both sides.

Upgrade

To update to this release:

npm i agents@latest

yarn add agents@latest

pnpm add agents@latest

bun add agents@latest

Agents, Workers - Agents SDK adds background sub-agents and a unified turn entry point

Fri, 26 Jun 2026 00:00:00 GMT

The latest release of the Agents SDK ↗ makes it easier to run long work in the background, drive turns through one entry point, and keep chat agents working through deploys, evictions, and reconnects.

This release adds first-class detached (background) sub-agent runs with live progress and durable milestones, a single runTurn turn-admission entry point, and a large round of recovery and reliability fixes that continue converging @cloudflare/think and @cloudflare/ai-chat onto one model.

Background sub-agents with progress and milestones

runAgentTool can now dispatch a sub-agent without blocking the calling turn. A detached run returns a handle immediately and is owned by a durable, eviction-surviving backbone instead of being abandoned when the dispatching turn ends.

class OrdersAgent extends Think {
	async startImport(input) {
		// Fire-and-forget, or wire a durable completion callback
		// (by method name, like schedule()):
		await this.runAgentTool(ImportAgent, {
			input,
			detached: { onFinish: "onImportDone", maxBudgetMs: 60 * 60 * 1000 },
		});
	}

	// result.status: "completed" | "error" | "aborted" | "interrupted"
	async onImportDone(run, result) {}
}

class OrdersAgent extends Think {
	async startImport(input) {
		// Fire-and-forget, or wire a durable completion callback
		// (by method name, like schedule()):
		await this.runAgentTool(ImportAgent, {
			input,
			detached: { onFinish: "onImportDone", maxBudgetMs: 60 * 60 * 1000 },
		});
	}

	// result.status: "completed" | "error" | "aborted" | "interrupted"
	async onImportDone(run, result) {}
}

Highlights:

Durable, exactly-once-on-the-happy-path completion via a warm fast path plus a self-scheduling reconcile backbone that survives eviction and deploys.
Bounded. An absolute maxBudgetMs ceiling (default 24h) and cancelAgentTool(runId) keep abandoned runs from holding a concurrency slot forever.
detached: { notify: true } lets a finished background run inject a message back into the chat so the model reacts to the result — no hand-wired onFinish needed.

Sub-agents can also report mid-run progress that rides their own turn stream back to the parent's connected clients:

// Inside the child sub-agent:
await this.reportProgress({
	fraction: 0.6,
	phase: "deploying",
	message: "Generating menu page…",
});

// Inside the child sub-agent:
await this.reportProgress({
	fraction: 0.6,
	phase: "deploying",
	message: "Generating menu page…",
});

Progress surfaces on AgentToolRunState.progress via useAgentToolEvents, so a background-runs tray can render a live bar without drilling in, and the latest snapshot is persisted for inspection after eviction. Naming a milestone promotes a signal to a durable, replayable row, and detached: { onMilestones } can surface a milestone as a synthetic chat message ("narrate" for a cheap status line, or "react" to drive a model turn).

One entry point for turns: `runTurn`

@cloudflare/think adds a public runTurn(options) facade that unifies turn admission behind a single mode:

await this.runTurn({ mode: "wait", messages }); // saveMessages / continueLastTurn
await this.runTurn({ mode: "submit", messages }); // durable submitMessages
await this.runTurn({ mode: "stream", messages }); // chat()

await this.runTurn({ mode: "wait", messages }); // saveMessages / continueLastTurn
await this.runTurn({ mode: "submit", messages }); // durable submitMessages
await this.runTurn({ mode: "stream", messages }); // chat()

stream mode accepts array and function inputs to match wait mode, and all entry points now route through a shared internal admission path that throws a clear error on nested blocking admissions that previously could deadlock.

Recovery and reliability

A large part of this release continues hardening recovery and converging @cloudflare/think and @cloudflare/ai-chat onto one model:

Stream stall watchdog. AIChatAgent can detect and recover from a hung model/transport stream via the opt-in chatStreamStallTimeoutMs watchdog. With chatRecovery enabled the stall routes into the same bounded-recovery machinery a deploy or eviction uses; otherwise it surfaces as a terminal stream error so the spinner clears.
Interrupted tool-call repair. AIChatAgent now repairs a transcript with a dead server-tool call before re-entering inference (parity with @cloudflare/think), so a recovered turn no longer fails with AI_MissingToolResultsError. An overridable repairInterruptedToolPart(part) hook lets apps customize the repaired shape.
Stuck status after reconnect. Fixed AI SDK status getting stuck when a reconnect races a turn that has been accepted but has not started streaming yet, so the UI now renders the in-flight turn instead of settling on ready.
Live "recovering…" on connect. AIChatAgent now replays the recovering status to a client that connects mid-recovery, so useAgentChat's isRecovering reflects in-progress recovery immediately instead of appearing frozen.
Terminal connection failures. The client stops reconnecting on terminal WebSocket close events and exposes them via connectionError / onConnectionError on AgentClient, useAgent, and useAgentChat.
Agent-tool child recovery. A healthy long-running sub-agent run is no longer abandoned as interrupted after a deploy (both @cloudflare/think and AIChatAgent).
Workflows from sub-agent facets. Agent Workflows can now start from sub-agent facets, with callbacks and Workflow RPC routed back to the originating facet.
Plus forward-progress crediting convergence, broadcast-first give-up ordering, an event-driven auto-continuation barrier, and structured row-size compaction in AIChatAgent.

Other improvements

Shared chat React core. A new agents/chat/react entry exposes useAgentChat, transport helpers, and shared wire types, with syncMessagesToServer for server-authoritative transcript storage. @cloudflare/think/react and @cloudflare/ai-chat/react are now thin wrappers over it.
Optional ai peer. The root agents and @cloudflare/codemode runtimes no longer reference AI SDK types, so they bundle without ai / zod installed; AI-specific entry points still require the peer when imported. just-bash likewise moves to an optional peer used only by the skills bash runner.
Code Mode. The default DynamicWorkerExecutor timeout increases from 30s to 60s, executions now dispose the dynamically-loaded Worker and its RPC stub after each run (fixing a flaky isolate-shutdown assertion), connector imports are cleaned up, and the outer MCP tool-call context is passed to openApiMcpServer request callbacks.
Voice. Voice turns now support AI SDK fullStream responses (and warn when textStream is used).
MCP. McpAgent server-to-client requests can now be sent from callbacks that do not inherit the agent's async context, including callbacks reached through Worker Loader RPC.
Experimental: server actions and channels. This release lays groundwork for guarded server actions (action() / getActions() with a durable replay ledger and approvals) and a unified channels surface (configureChannels(), deliverNotice()). Both are experimental and their APIs may change, so we don't recommend depending on them yet.

Upgrade

To update to the latest version:

npm i agents@latest @cloudflare/think@latest @cloudflare/ai-chat@latest @cloudflare/codemode@latest @cloudflare/voice@latest

yarn add agents@latest @cloudflare/think@latest @cloudflare/ai-chat@latest @cloudflare/codemode@latest @cloudflare/voice@latest

pnpm add agents@latest @cloudflare/think@latest @cloudflare/ai-chat@latest @cloudflare/codemode@latest @cloudflare/voice@latest

bun add agents@latest @cloudflare/think@latest @cloudflare/ai-chat@latest @cloudflare/codemode@latest @cloudflare/voice@latest

Refer to the Think documentation, Code Mode documentation, and Agents documentation for more information.

Agents, Workers - Agents SDK improves browser automation, code execution, and recovery

Tue, 16 Jun 2026 00:00:00 GMT

The latest release of the Agents SDK ↗ makes it easier to build agents that can safely interact with real systems and keep working through interruptions.

Agents can now browse websites through Browser Run, write code against external tools through Code Mode, use client-provided tools when delegating to Think sub-agents, and recover more reliably from deploys, Durable Object evictions, and connection churn.

Safer browser automation

Agents can now use Browser Run through a single durable browser_execute tool. Instead of choosing from a fixed list of actions, the model writes code against the Chrome DevTools Protocol (CDP) and can inspect pages, capture screenshots, read rendered content, debug frontend behavior, and interact with live browser sessions.

const browserTools = createBrowserTools({
	ctx: this.ctx,
	browser: this.env.BROWSER,
	loader: this.env.LOADER,
	session: { mode: "dynamic" },
});

const browserTools = createBrowserTools({
	ctx: this.ctx,
	browser: this.env.BROWSER,
	loader: this.env.LOADER,
	session: { mode: "dynamic" },
});

Browser sessions can be one-time, reused, or promoted from one-time to persistent during a run. This is useful when an agent needs a human to log in, complete MFA, or approve a sensitive action. The run can pause, keep the same tabs and cookies, and resume after approval.

The browser tools also add Live View URLs, optional session recording, and quick actions such as browser_markdown, browser_extract, browser_links, and browser_scrape for one-shot browsing tasks.

Resumable code execution with approvals

Code Mode now uses createCodemodeRuntime, connectors, and a durable execution log. This lets you give a model one codemode tool instead of a large prompt full of tool definitions. The model can discover the capabilities it needs, write code against typed globals, and reuse saved snippets.

const runtime = createCodemodeRuntime({
	ctx: this.ctx,
	executor: new DynamicWorkerExecutor({ loader: this.env.LOADER }),
	connectors: [new GithubConnector(this.ctx, this.env, connection)],
});

const result = streamText({
	model,
	messages,
	tools: { codemode: runtime.tool() },
});

const runtime = createCodemodeRuntime({
	ctx: this.ctx,
	executor: new DynamicWorkerExecutor({ loader: this.env.LOADER }),
	connectors: [new GithubConnector(this.ctx, this.env, connection)],
});

const result = streamText({
	model,
	messages,
	tools: { codemode: runtime.tool() },
});

When the code reaches an approval-gated action, the runtime pauses execution and returns a pending approval. After approval, completed calls replay from the durable log, the approved action runs, and the same code continues. This makes it practical to build agents that create issues, update external systems, or perform other side effects without custom pause-and-resume logic for every tool.

Better Think delegation

Think sub-agents can now use client-defined tools over the RPC chat() path. A parent agent can pass tool schemas with clientTools and resolve tool calls through onClientToolCall. This lets delegated agents use caller-provided capabilities without requiring a browser WebSocket.

await child.chat(message, callback, {
	signal,
	clientTools: [
		{
			name: "get_user_timezone",
			description: "Get the caller's timezone",
			parameters: { type: "object" },
		},
	],
	onClientToolCall: async ({ toolName, input }) => {
		return runClientTool(toolName, input);
	},
});

await child.chat(message, callback, {
	signal,
	clientTools: [
		{
			name: "get_user_timezone",
			description: "Get the caller's timezone",
			parameters: { type: "object" },
		},
	],
	onClientToolCall: async ({ toolName, input }) => {
		return runClientTool(toolName, input);
	},
});

Think Workflows also improve step.prompt(). A prompt step now runs a full agentic turn before returning structured output, so the agent can call tools before producing the typed result. This makes Workflow steps more useful for durable triage, research, and approval flows.

The unified Think execute tool can also include cdp.* browser capabilities alongside state.* and tools.* when Browser Run is bound.

Voice output device selection

Voice clients can route assistant audio to a specific output device. Use outputDeviceId with useVoiceAgent, or call client.setOutputDevice() from the framework-agnostic client.

const voice = useVoiceAgent({
	agent: "MyVoiceAgent",
	outputDeviceId: selectedSpeakerId,
});

const voice = useVoiceAgent({
	agent: "MyVoiceAgent",
	outputDeviceId: selectedSpeakerId,
});

Browsers without speaker-selection support continue playing through the default output device and report a non-fatal outputDeviceError.

Reliability fixes

This release includes several fixes for production agents:

useAgent and AgentClient handle WebSocket replacement more reliably during reconnects and configuration changes.
Chat stream replay is more reliable after reconnects, deploys, and provider errors.
Fiber recovery continues across multi-pass scans and backs off when recovery hooks keep failing.
Agent teardown continues even when the request that started teardown is canceled.
Large session histories use byte-budgeted reads to reduce memory pressure during startup.

Upgrade

To update to the latest version:

npm i agents@latest @cloudflare/think@latest @cloudflare/codemode@latest @cloudflare/ai-chat@latest @cloudflare/voice@latest

yarn add agents@latest @cloudflare/think@latest @cloudflare/codemode@latest @cloudflare/ai-chat@latest @cloudflare/voice@latest

pnpm add agents@latest @cloudflare/think@latest @cloudflare/codemode@latest @cloudflare/ai-chat@latest @cloudflare/voice@latest

bun add agents@latest @cloudflare/think@latest @cloudflare/codemode@latest @cloudflare/ai-chat@latest @cloudflare/voice@latest

Refer to the Code Mode documentation, Browser tools documentation, Think tools documentation, and Voice documentation for more information.

Workers, Agents, Workers AI - Introducing GLM-5.2 on Workers AI

Tue, 16 Jun 2026 00:00:00 GMT

We are excited to announce GLM-5.2 on Workers AI, Z.ai's flagship agentic coding model.

@cf/zai-org/glm-5.2 is a text generation model built for agentic coding workflows. With function calling and reasoning support, it can handle long codebases, multi-step planning, and tool-augmented agents.

Key features and use cases:

Agentic coding: Designed for autonomous coding tasks, long-horizon planning, and complex software engineering workflows
Large context window: GLM-5.2 supports up to a 1,048,576 token context window. Workers AI is launching the model with a 262,144 token context window and plans to increase this in the future
Function calling: Build agents that invoke tools and APIs across multiple conversation turns
Reasoning: Tackles complex problem-solving and step-by-step reasoning tasks

Use GLM-5.2 through the Workers AI binding (env.AI.run()), the REST API at /run or /v1/chat/completions, or AI Gateway.

Pricing is available on the model page or pricing page.

Agents, Workers - Agents SDK v0.14.0: Agent Skills, messengers, scheduled tasks, Workflows, and hardened chat recovery

Tue, 02 Jun 2026 00:00:00 GMT

The latest release of the Agents SDK ↗ adds four new ways to build with @cloudflare/think: on-demand Agent Skills, chat messengers (starting with Telegram), declarative scheduled tasks, and durable reasoning steps inside Workflows. This release also significantly hardens durable chat recovery, so turns reliably ride through deploys, evictions, and stalled model streams in production.

Agent Skills (experimental)

Give an agent a catalog of on-demand instructions, resources, and scripts. A skill source adds a catalog to the system prompt, and the model activates a skill only when a task matches — so a large library of capabilities does not bloat every prompt.

import { Think, skills } from "@cloudflare/think";
import bundledSkills from "agents:skills";

export class SkillsAgent extends Think {
	getSkills() {
		return [
			bundledSkills,
			skills.r2(this.env.SKILLS_BUCKET, { prefix: "skills/" }),
		];
	}
}

import { Think, skills } from "@cloudflare/think";
import bundledSkills from "agents:skills";

export class SkillsAgent extends Think<Env> {
	getSkills() {
		return [
			bundledSkills,
			skills.r2(this.env.SKILLS_BUCKET, { prefix: "skills/" }),
		];
	}
}

The agents:skills import bundles a local ./skills directory through the Agents Vite plugin (one directory per skill, each with a SKILL.md). Skills can also load from R2 or a manifest. When skills are available, Think exposes activate_skill, read_skill_resource, and an optional run_skill_script tool. Skill loading is resilient: a duplicate or failing source is skipped with a warning instead of breaking the agent.

Agent Skills are experimental, and script execution in particular is early. The API may change in a future release. We would love your feedback — tell us what you are building and what is missing in the Agents repository ↗.

Messengers

Connect a Think agent directly to a chat platform. Think owns the webhook route, conversation routing, durable reply fiber, and streamed delivery back to the provider. Telegram ships as the first provider.

import { Think } from "@cloudflare/think";
import {
	defineMessengers,
	ThinkMessengerStateAgent,
} from "@cloudflare/think/messengers";
import telegramMessenger from "@cloudflare/think/messengers/telegram";

export { ThinkMessengerStateAgent };

export class SupportAgent extends Think {
	getMessengers() {
		return defineMessengers({
			telegram: telegramMessenger({
				token: this.env.TELEGRAM_BOT_TOKEN,
				userName: "support_bot",
				secretToken: this.env.TELEGRAM_WEBHOOK_SECRET_TOKEN,
			}),
		});
	}
}

import { Think } from "@cloudflare/think";
import {
	defineMessengers,
	ThinkMessengerStateAgent,
} from "@cloudflare/think/messengers";
import telegramMessenger from "@cloudflare/think/messengers/telegram";

export { ThinkMessengerStateAgent };

export class SupportAgent extends Think<Env> {
	getMessengers() {
		return defineMessengers({
			telegram: telegramMessenger({
				token: this.env.TELEGRAM_BOT_TOKEN,
				userName: "support_bot",
				secretToken: this.env.TELEGRAM_WEBHOOK_SECRET_TOKEN,
			}),
		});
	}
}

Each Chat SDK thread maps to its own Think sub-agent by default, so group chats and direct messages do not share memory. Multiple bots, custom conversation routing, and custom providers are all supported.

Scheduled tasks

Declare recurring, timezone-aware prompts and handlers with a typed domain-specific language (DSL). Think reconciles the declarations on startup and re-arms the next occurrence after each run, backed by durable idempotent submissions.

import { Think, defineScheduledTasks } from "@cloudflare/think";

export class DigestAgent extends Think {
	getScheduledTasks() {
		return defineScheduledTasks({
			weeklyCommitReport: {
				schedule: "every week on monday at 09:00",
				prompt:
					"Compile my GitHub commits for the last week and summarize them.",
			},
			workout: {
				schedule: "every day at 08:00 in Europe/London",
				prompt: "Start my workout.",
			},
		});
	}
}

import { Think, defineScheduledTasks } from "@cloudflare/think";

export class DigestAgent extends Think<Env> {
	getScheduledTasks() {
		return defineScheduledTasks({
			weeklyCommitReport: {
				schedule: "every week on monday at 09:00",
				prompt:
					"Compile my GitHub commits for the last week and summarize them.",
			},
			workout: {
				schedule: "every day at 08:00 in Europe/London",
				prompt: "Start my workout.",
			},
		});
	}
}

Think Workflows

Run a model-driven reasoning step inside a Cloudflare Workflow with ThinkWorkflow and step.prompt(), with durable typed structured output, long waits, and approval gates.

import { z } from "zod";
import { ThinkWorkflow } from "@cloudflare/think/workflows";

const draftSchema = z.object({
	title: z.string(),
	summary: z.string(),
	labels: z.array(z.string()),
});

export class TriageWorkflow extends ThinkWorkflow {
	async run(event, step) {
		const draft = await step.prompt("triage-issue", {
			prompt: `Triage issue #${event.payload.issueNumber}`,
			output: draftSchema,
			timeout: "3 days",
		});

		await step.do("apply-labels", async () => {
			await this.agent.applyLabels(draft.labels);
		});
	}
}

import { z } from "zod";
import { ThinkWorkflow } from "@cloudflare/think/workflows";
import type { ThinkWorkflowStep } from "@cloudflare/think/workflows";
import type { AgentWorkflowEvent } from "agents/workflows";

const draftSchema = z.object({
	title: z.string(),
	summary: z.string(),
	labels: z.array(z.string()),
});

export class TriageWorkflow extends ThinkWorkflow<TriageAgent, Params> {
	async run(event: AgentWorkflowEvent<Params>, step: ThinkWorkflowStep) {
		const draft = await step.prompt("triage-issue", {
			prompt: `Triage issue #${event.payload.issueNumber}`,
			output: draftSchema,
			timeout: "3 days",
		});

		await step.do("apply-labels", async () => {
			await this.agent.applyLabels(draft.labels);
		});
	}
}

Production hardening for durable chat recovery

Durable chat turns have always been designed to survive a mid-turn deploy or Durable Object eviction. This release is a major hardening pass on that machinery for production.

Better recovery during deploys. Turns now ride through continuous deploys and evictions without losing completed work or re-running tools that already ran.
A live "recovering…" signal. useAgentChat exposes a new isRecovering flag, so a recovering turn shows progress instead of looking frozen. Most UIs render isStreaming || isRecovering as "busy".
Stalled streams recover. Set chatStreamStallTimeoutMs to route a hung provider stream into the same recovery path instead of leaving an infinite spinner.
Sub-agents re-attach. On parent recovery, an in-flight agentTool() child is re-attached to its result rather than abandoned and re-run, so long-running children no longer lose work under deploys.

MCP transport improvements

Resumable streams — In-flight tool calls over Server-Sent Events (SSE) survive a dropped connection. Clients reconnect with Last-Event-ID and replay anything they missed.
Readable server IDs — addMcpServer accepts an optional id, so tools surface as readable keys (for example tool_github_create_pull_request) instead of opaque connection IDs.
Better handling of concurrent requests — Overlapping JSON-RPC requests are now correctly correlated to their responses across the HTTP and RPC transports.

Other improvements

Compaction — A Session's tokenCounter now also drives the compaction boundary decision ("what to compress"), not just the fire/no-fire trigger.
@cloudflare/worker-bundler — Adds a virtualModules option to createWorker to provide in-memory module source during bundling.
Client-tool continuations — Parallel tool results now coalesce into a single continuation, immediate resume requests attach to the pending continuation, and server-side needsApproval continuations resume reliably after approval.

Upgrade

To update to the latest version:

npm i agents@latest @cloudflare/think@latest @cloudflare/ai-chat@latest

yarn add agents@latest @cloudflare/think@latest @cloudflare/ai-chat@latest

pnpm add agents@latest @cloudflare/think@latest @cloudflare/ai-chat@latest

bun add agents@latest @cloudflare/think@latest @cloudflare/ai-chat@latest

Refer to the Agents API reference and Chat agents documentation for more information.

Agents - Share sandbox previews through Cloudflare Tunnel

Fri, 29 May 2026 00:00:00 GMT

Sandboxes can expose a service running inside the container on a public preview URL through the sandbox.tunnels namespace. The SDK uses cloudflared inside the sandbox so you can share a running service without configuring exposePort() or a custom domain.

By default, sandbox.tunnels.get(port) creates a quick tunnel ↗ on a zero-config *.trycloudflare.com URL — no Cloudflare account, DNS record, or custom domain required. This is perfect for quick development and for .workers.dev deployments.

import { getSandbox } from "@cloudflare/sandbox";

const sandbox = getSandbox(env.Sandbox, "my-sandbox");
await sandbox.startProcess("python -m http.server 8080");

const tunnel = await sandbox.tunnels.get(8080);
console.log(tunnel.url); // → https://random-words-here.trycloudflare.com

import { getSandbox } from "@cloudflare/sandbox";

const sandbox = getSandbox(env.Sandbox, "my-sandbox");
await sandbox.startProcess("python -m http.server 8080");

const tunnel = await sandbox.tunnels.get(8080);
console.log(tunnel.url); // → https://random-words-here.trycloudflare.com

Named tunnels

For more control you can create a named tunnel through sandbox.tunnels.get(port, { name }). A named tunnel binds a hostname (<name>.<your-zone>) backed by a Cloudflare Tunnel and a CNAME record on your zone resulting in something like https://my-app-preview.example.com ↗.

Unlike quick tunnels, which generate a new random URL each time, a named tunnel produces a persistent URL that survives container restarts. This makes named tunnels suitable for production use cases where you want control over the tunnel and it's origin.

const tunnel = await sandbox.tunnels.get(8080, { name: "my-app-preview" });
console.log(tunnel.url); // → https://my-app-preview.example.com

const tunnel = await sandbox.tunnels.get(8080, { name: "my-app-preview" });
console.log(tunnel.url); // → https://my-app-preview.example.com

Calling sandbox.destroy() tears down the Cloudflare Tunnel and the associated DNS record alongside the container, so you do not leave dangling tunnels or records behind.

Upgrade

To update to the latest version:

npm i @cloudflare/sandbox@latest

yarn add @cloudflare/sandbox@latest

pnpm add @cloudflare/sandbox@latest

bun add @cloudflare/sandbox@latest

For full API details, refer to the Sandbox tunnels reference.

Agents, Workers - Agents SDK v0.12.4: chat recovery, routing retries, durable Think submissions, and Voice connection control

Wed, 13 May 2026 00:00:00 GMT

The latest release of the Agents SDK ↗ brings more reliable chat recovery, fixes Agent state synchronization during reconnects, adds durable submissions for Think, exposes routing retry configuration, and adds connection control for Voice agents.

Chat recovery improvements

@cloudflare/ai-chat now keeps server turns running when a browser or client stream is interrupted. This is useful for long-running AI responses where users refresh the page, close a tab, or temporarily lose connection. Calling stop() still cancels the server turn.

Set cancelOnClientAbort: true if browser or client aborts should also cancel the server turn:

const chat = useAgentChat({
	agent: "assistant",
	name: "user-123",
	cancelOnClientAbort: true,
});

const chat = useAgentChat({
	agent: "assistant",
	name: "user-123",
	cancelOnClientAbort: true,
});

Notable bug fixes:

Chat stream resume negotiation no longer throws when replay races with a closed WebSocket connection.
Recovered chat continuations no longer leave useAgentChat stuck in a streaming state when the original socket disconnects before a terminal response.
Approval auto-continuation preserves reasoning parts and persists continuation reasoning in the final message.
isServerStreaming now resets correctly when a resumed stream moves from the fallback observer path to a transport-owned stream.

Agent state and routing fixes

agents@0.12.4 prevents duplicate initial state frames during WebSocket connection setup. This avoids stale initial state messages overwriting state updates already sent by the client.

Agent recovery is also more reliable when tool calls span a Durable Object restart. Recovery now defers user finish hooks until after agent startup and isolates hook failures, so one failed hook does not block other recovered runs from finalizing.

getAgentByName() now supports routingRetry for transient Durable Object routing failures:

import { getAgentByName } from "agents";

const agent = await getAgentByName(env.AssistantAgent, "user-123", {
	routingRetry: {
		maxAttempts: 3,
	},
});

import { getAgentByName } from "agents";

const agent = await getAgentByName(env.AssistantAgent, "user-123", {
	routingRetry: {
		maxAttempts: 3,
	},
});

Durable Think submissions

@cloudflare/think now supports durable programmatic submissions. submitMessages() provides durable acceptance, idempotent retries, status inspection, cancellation, and cleanup for server-driven turns that should continue after the caller returns.

Think.chat() RPC turns now run inside chat recovery fibers and persist their stream chunks. Interrupted sub-agent turns can recover partial output instead of starting over.

ChatOptions.tools has been removed from the TypeScript API. Define durable tools on the child agent or use agent tools for orchestration. Runtime options.tools values passed by legacy callers are ignored with a warning.

Think message pruning behavior change

@cloudflare/think no longer applies pruneMessages({ toolCalls: "before-last-2-messages" }) to model context by default. The previous default could strip client-side tool results from longer multi-turn flows.

truncateOlderMessages still runs as before, so context cost remains bounded. Subclasses that relied on the old aggressive pruning can opt back in from beforeTurn:

import { Think } from "@cloudflare/think";
import { pruneMessages } from "ai";

export class MyAgent extends Think {
	beforeTurn(ctx) {
		return {
			messages: pruneMessages({
				messages: ctx.messages,
				toolCalls: "before-last-2-messages",
			}),
		};
	}
}

import { Think } from "@cloudflare/think";
import { pruneMessages } from "ai";

export class MyAgent extends Think<Env> {
	beforeTurn(ctx) {
		return {
			messages: pruneMessages({
				messages: ctx.messages,
				toolCalls: "before-last-2-messages",
			}),
		};
	}
}

Voice agent connection control

@cloudflare/voice adds an enabled option to useVoiceAgent. React apps can now delay creating and connecting a VoiceClient until prerequisites such as capability tokens are ready.

const voice = useVoiceAgent({
	agent: "MyVoiceAgent",
	enabled: Boolean(token),
});

const voice = useVoiceAgent({
	agent: "MyVoiceAgent",
	enabled: Boolean(token),
});

This release also fixes Workers AI speech-to-text session edge cases and withVoice text streaming from AI SDK textStream responses.

Other improvements

Streamable HTTP routing — Server-to-client requests now route through the originating POST stream when no standalone SSE stream is available.
Structured tool output — Tool output shapes are preserved when truncating older messages or oversized persisted rows.
Non-chat Think tool steps — Think agent-tool children can complete without emitting assistant text and can return structured output through getAgentToolOutput.
Sub-agent schedules — Stale sub-agent schedule rows are pruned when their owning facet registry entry no longer exists.
@cloudflare/codemode — Adds a browser-safe export with an iframe sandbox executor and resolves OpenAPI specs inside the sandbox to avoid Worker Loader RPC size limits.

Upgrade

To update to the latest version:

npm i agents@latest @cloudflare/ai-chat@latest @cloudflare/think@latest @cloudflare/voice@latest

Refer to the Agents API reference and Chat agents documentation for more information.

Agents - Agent Lee adds Write Operations and Generative UI

Wed, 15 Apr 2026 00:00:00 GMT

Agent Lee adds Write Operations and Generative UI

We are excited to announce two major capability upgrades for Agent Lee, the AI co-pilot built directly into the Cloudflare dashboard. Agent Lee is designed to understand your specific account configuration, and with this release, it moves from a passive advisor to an active assistant that can help you manage your infrastructure and visualize your data through natural language.

Take action with Write Operations

Agent Lee can now perform changes on your behalf across your Cloudflare account. Whether you need to update DNS records, modify SSL/TLS settings, or configure Workers routes, you can simply ask.

To ensure security and accuracy, every write operation requires explicit user approval. Before any change is committed, Agent Lee will present a summary of the proposed action in plain language. No action is taken until you select Confirm, and this approval requirement is enforced at the infrastructure level to prevent unauthorized changes.

Example requests:

"Add an A record for blog.example.com pointing to 192.0.2.10."
"Enable Always Use HTTPS on my zone."
"Set the SSL mode for example.com to Full (strict)."

Visualize data with Generative UI

Understanding your traffic and security trends is now as easy as asking a question. Agent Lee now features Generative UI, allowing it to render inline charts and structured data visualizations directly within the chat interface using your actual account telemetry.

Example requests:

"Show me a chart of my traffic over the last 7 days."
"What does my error rate look like for the past 24 hours?"
"Graph my cache hit rate for example.com this week."

Availability

These features are currently available in Beta for all users on the Free plan. To get started, log in to the Cloudflare dashboard ↗ and select Ask AI in the upper right corner.

To learn more about how to interact with your account using AI, refer to the Agent Lee documentation.

Containers, Agents - Secure credential injection and dynamic egress policies for Sandboxes

Mon, 13 Apr 2026 00:00:00 GMT

Outbound Workers for Sandboxes and Containers now support zero-trust credential injection, TLS interception, allow/deny lists, and dynamic per-instance egress policies. These features give platforms running agentic workloads full control over what leaves the sandbox, without exposing secrets to untrusted workloads, like user-generated code or coding agents.

Credential injection

Because outbound handlers run in the Workers runtime, outside the sandbox, they can hold secrets the sandbox never sees. A sandboxed workload can make a plain request, and credentials are transparently attached before a request is forwarded upstream.

For instance, you could run an agent in a sandbox and ensure that any requests it makes to Github are authenticated. But it will never be able to access the credentials:

export class MySandbox extends Sandbox {}

MySandbox.outboundByHost = {
	"github.com": (request: Request, env: Env, ctx: OutboundHandlerContext) => {
		const requestWithAuth = new Request(request);
		requestWithAuth.headers.set("x-auth-token", env.SECRET);
		return fetch(requestWithAuth);
	},
};

You can easily inject unique credentials for different instances by using ctx.containerId:

MySandbox.outboundByHost = {
	"my-internal-vcs.dev": async (
		request: Request,
		env: Env,
		ctx: OutboundHandlerContext,
	) => {
		const authKey = await env.KEYS.get(ctx.containerId);

		const requestWithAuth = new Request(request);
		requestWithAuth.headers.set("x-auth-token", authKey);
		return fetch(requestWithAuth);
	},
};

No token is ever passed into the sandbox. You can rotate secrets in the Worker environment and every request will pick them up immediately.

TLS interception

Outbound Workers now intercept HTTPS traffic. A unique ephemeral certificate authority (CA) and private key are created for each sandbox instance. The CA is placed into the sandbox and trusted by default. The ephemeral private key never leaves the container runtime sidecar process and is never shared across instances.

With TLS interception active, outbound Workers can act as a transparent proxy for both HTTP and HTTPS traffic.

Allow and deny hosts

Easily filter outbound traffic with allowedHosts and deniedHosts. When allowedHosts is set, it becomes a deny-by-default allowlist. Both properties support glob patterns.

export class MySandbox extends Sandbox {
	allowedHosts = ["github.com", "npmjs.org"];
}

Dynamic outbound handlers

Define named outbound handlers then apply or remove them at runtime using setOutboundHandler() or setOutboundByHost(). This lets you change egress policy for a running sandbox without restarting it.

export class MySandbox extends Sandbox {}

MySandbox.outboundHandlers = {
	allowHosts: async (req: Request, env: Env, ctx: OutboundHandlerContext ) => {
		const url = new URL(req.url);
		if (ctx.params.allowedHostnames.includes(url.hostname)) {
			return fetch(req);
		}
		return new Response(null, { status: 403 });
	},

	noHttp: async () => {
		return new Response(null, { status: 403 });
	},
};

Apply handlers programmatically from your Worker:

const sandbox = getSandbox(env.Sandbox, userId);

// Open network for setup
await sandbox.setOutboundHandler("allowHosts", {
	allowedHostnames: ["github.com", "npmjs.org"],
});
await sandbox.exec("npm install");

// Lock down after setup
await sandbox.setOutboundHandler("noHttp");

Handlers accept params, so you can customize behavior per instance without defining separate handler functions.

Get started

Upgrade to @cloudflare/containers@0.3.0 or @cloudflare/sandbox@0.8.9 to use these features.

For more details, refer to Sandbox outbound traffic and Container outbound traffic.

Agents, Workers - Agents SDK v0.8.0: readable state, idempotent schedules, typed AgentClient, and Zod 4

Mon, 23 Mar 2026 00:00:00 GMT

The latest release of the Agents SDK ↗ exposes agent state as a readable property, prevents duplicate schedule rows across Durable Object restarts, brings full TypeScript inference to AgentClient, and migrates to Zod 4.

Readable `state` on `useAgent` and `AgentClient`

Both useAgent (React) and AgentClient (vanilla JS) now expose a state property that reflects the current agent state. Previously, reading state required manually tracking it through the onStateUpdate callback.

React (useAgent)

const agent = useAgent({
	agent: "game-agent",
	name: "room-123",
});

// Read state directly — no separate useState + onStateUpdate needed
return <div>Score: {agent.state?.score}</div>;

// Spread for partial updates
agent.setState({ ...agent.state, score: (agent.state?.score ?? 0) + 10 });

const agent = useAgent<GameAgent, GameState>({
	agent: "game-agent",
	name: "room-123",
});

// Read state directly — no separate useState + onStateUpdate needed
return <div>Score: {agent.state?.score}</div>;

// Spread for partial updates
agent.setState({ ...agent.state, score: (agent.state?.score ?? 0) + 10 });

agent.state is reactive — the component re-renders when state changes from either the server or a client-side setState() call.

Vanilla JS (AgentClient)

const client = new AgentClient({
	agent: "game-agent",
	name: "room-123",
	host: "your-worker.workers.dev",
});

client.setState({ score: 100 });
console.log(client.state); // { score: 100 }

const client = new AgentClient<GameAgent>({
	agent: "game-agent",
	name: "room-123",
	host: "your-worker.workers.dev",
});

client.setState({ score: 100 });
console.log(client.state); // { score: 100 }

State starts as undefined and is populated when the server sends the initial state on connect (from initialState) or when setState() is called. Use optional chaining (agent.state?.field) for safe access. The onStateUpdate callback continues to work as before — the new state property is additive.

Idempotent `schedule()`

schedule() now supports an idempotent option that deduplicates by (type, callback, payload), preventing duplicate rows from accumulating when called in places that run on every Durable Object restart such as onStart().

Cron schedules are idempotent by default. Calling schedule("0 * * * *", "tick") multiple times with the same callback, expression, and payload returns the existing schedule row instead of creating a new one. Pass { idempotent: false } to override.

Delayed and date-scheduled types support opt-in idempotency:

import { Agent } from "agents";

class MyAgent extends Agent {
	async onStart() {
		// Safe across restarts — only one row is created
		await this.schedule(60, "maintenance", undefined, { idempotent: true });
	}
}

import { Agent } from "agents";

class MyAgent extends Agent {
	async onStart() {
		// Safe across restarts — only one row is created
		await this.schedule(60, "maintenance", undefined, { idempotent: true });
	}
}

Two new warnings help catch common foot-guns:

Calling schedule() inside onStart() without { idempotent: true } emits a console.warn with actionable guidance (once per callback; skipped for cron and when idempotent is set explicitly).
If an alarm cycle processes 10 or more stale one-shot rows for the same callback, the SDK emits a console.warn and a schedule:duplicate_warning diagnostics channel event.

Typed `AgentClient` with `call` inference and `stub` proxy

AgentClient now accepts an optional agent type parameter for full type inference on RPC calls, matching the typed experience already available with useAgent.

const client = new AgentClient({
	agent: "my-agent",
	host: window.location.host,
});

// Typed call — method name autocompletes, args and return type inferred
const value = await client.call("getValue");

// Typed stub — direct RPC-style proxy
await client.stub.getValue();
await client.stub.add(1, 2);

const client = new AgentClient<MyAgent>({
	agent: "my-agent",
	host: window.location.host,
});

// Typed call — method name autocompletes, args and return type inferred
const value = await client.call("getValue");

// Typed stub — direct RPC-style proxy
await client.stub.getValue();
await client.stub.add(1, 2);

State is automatically inferred from the agent type, so onStateUpdate is also typed:

const client = new AgentClient({
	agent: "my-agent",
	host: window.location.host,
	onStateUpdate: (state) => {
		// state is typed as MyAgent's state type
	},
});

const client = new AgentClient<MyAgent>({
	agent: "my-agent",
	host: window.location.host,
	onStateUpdate: (state) => {
		// state is typed as MyAgent's state type
	},
});

Existing untyped usage continues to work without changes. The RPC type utilities (AgentMethods, AgentStub, RPCMethods) are now exported from agents/client for advanced typing scenarios. agents, @cloudflare/ai-chat, and @cloudflare/codemode now require zod ^4.0.0. Zod v3 is no longer supported.

`@cloudflare/ai-chat` fixes

Turn serialization — onChatMessage() and _reply() work is now queued so user requests, tool continuations, and saveMessages() never stream concurrently.
Duplicate messages on stop — Clicking stop during an active stream no longer splits the assistant message into two entries.
Duplicate messages after tool calls — Orphaned client IDs no longer leak into persistent storage.

`keepAlive()` and `keepAliveWhile()` are no longer experimental

keepAlive() now uses a lightweight in-memory ref count instead of schedule rows. Multiple concurrent callers share a single alarm cycle. The @experimental tag has been removed from both keepAlive() and keepAliveWhile().

`@cloudflare/codemode`: TanStack AI integration

A new entry point @cloudflare/codemode/tanstack-ai adds support for TanStack AI's ↗ chat() as an alternative to the Vercel AI SDK's streamText():

import {
	createCodeTool,
	tanstackTools,
} from "@cloudflare/codemode/tanstack-ai";
import { chat } from "@tanstack/ai";

const codeTool = createCodeTool({
	tools: [tanstackTools(myServerTools)],
	executor,
});

const stream = chat({ adapter, tools: [codeTool], messages });

import { createCodeTool, tanstackTools } from "@cloudflare/codemode/tanstack-ai";
import { chat } from "@tanstack/ai";

const codeTool = createCodeTool({
	tools: [tanstackTools(myServerTools)],
	executor,
});

const stream = chat({ adapter, tools: [codeTool], messages });

Upgrade

To update to the latest version:

npm i agents@latest @cloudflare/ai-chat@latest

Agents, Workers - @cloudflare/codemode v0.2.1: MCP barrel export, zero-dependency main entry point, and custom sandbox modules

Tue, 17 Mar 2026 00:00:00 GMT

The latest releases of @cloudflare/codemode ↗ add a new MCP barrel export, remove ai and zod as required peer dependencies from the main entry point, and give you more control over the sandbox.

New `@cloudflare/codemode/mcp` export

A new @cloudflare/codemode/mcp entry point provides two functions that wrap MCP servers with Code Mode:

codeMcpServer({ server, executor }) — wraps an existing MCP server with a single code tool where each upstream tool becomes a typed codemode.* method.
openApiMcpServer({ spec, executor, request }) — creates search and execute MCP tools from an OpenAPI spec with host-side request proxying and automatic $ref resolution.

import { codeMcpServer } from "@cloudflare/codemode/mcp";
import { DynamicWorkerExecutor } from "@cloudflare/codemode";

const executor = new DynamicWorkerExecutor({ loader: env.LOADER });

// Wrap an existing MCP server — all its tools become
// typed methods the LLM can call from generated code
const server = await codeMcpServer({ server: upstreamMcp, executor });

import { codeMcpServer } from "@cloudflare/codemode/mcp";
import { DynamicWorkerExecutor } from "@cloudflare/codemode";

const executor = new DynamicWorkerExecutor({ loader: env.LOADER });

// Wrap an existing MCP server — all its tools become
// typed methods the LLM can call from generated code
const server = await codeMcpServer({ server: upstreamMcp, executor });

Zero-dependency main entry point

Breaking change in v0.2.0: generateTypes and the ToolDescriptor / ToolDescriptors types have moved to @cloudflare/codemode/ai:

// Before
import { generateTypes } from "@cloudflare/codemode";

// After
import { generateTypes } from "@cloudflare/codemode/ai";

// Before
import { generateTypes } from "@cloudflare/codemode";

// After
import { generateTypes } from "@cloudflare/codemode/ai";

The main entry point (@cloudflare/codemode) no longer requires the ai or zod peer dependencies. It now exports:

Export	Description
`sanitizeToolName`	Sanitize tool names into valid JS identifiers
`normalizeCode`	Normalize LLM-generated code into async arrow functions
`generateTypesFromJsonSchema`	Generate TypeScript type definitions from plain JSON Schema
`jsonSchemaToType`	Convert a single JSON Schema to a TypeScript type string
`DynamicWorkerExecutor`	Sandboxed code execution via Dynamic Worker Loader
`ToolDispatcher`	RPC target for dispatching tool calls from sandbox to host

The ai and zod peer dependencies are now optional — only required when importing from @cloudflare/codemode/ai.

Custom sandbox modules

DynamicWorkerExecutor now accepts an optional modules option to inject custom ES modules into the sandbox:

const executor = new DynamicWorkerExecutor({
	loader: env.LOADER,
	modules: {
		"utils.js": `export function add(a, b) { return a + b; }`,
	},
});

// Sandbox code can then: import { add } from "utils.js"

const executor = new DynamicWorkerExecutor({
	loader: env.LOADER,
	modules: {
		"utils.js": `export function add(a, b) { return a + b; }`,
	},
});

// Sandbox code can then: import { add } from "utils.js"

Internal normalization and sanitization

DynamicWorkerExecutor now normalizes code and sanitizes tool names internally. You no longer need to call normalizeCode() or sanitizeToolName() before passing code and functions to execute().

Upgrade

npm i @cloudflare/codemode@latest

See the Code Mode documentation for the full API reference.

Agents - Real-time file watching in Sandboxes

Tue, 03 Mar 2026 00:00:00 GMT

Sandboxes now support real-time filesystem watching via sandbox.watch(). The method returns a Server-Sent Events ↗ stream backed by native inotify, so your Worker receives create, modify, delete, and move events as they happen inside the container.

`sandbox.watch(path, options)`

Pass a directory path and optional filters. The returned stream is a standard ReadableStream you can proxy directly to a browser client or consume server-side.

// Stream events to a browser client
const stream = await sandbox.watch("/workspace/src", {
	recursive: true,
	include: ["*.ts", "*.js"],
});

return new Response(stream, {
	headers: { "Content-Type": "text/event-stream" },
});

// Stream events to a browser client
const stream = await sandbox.watch("/workspace/src", {
	recursive: true,
	include: ["*.ts", "*.js"],
});

return new Response(stream, {
	headers: { "Content-Type": "text/event-stream" },
});

Server-side consumption with `parseSSEStream`

Use parseSSEStream to iterate over events inside a Worker without forwarding them to a client.

import { parseSSEStream } from "@cloudflare/sandbox";

const stream = await sandbox.watch("/workspace/src", { recursive: true });

for await (const event of parseSSEStream(stream)) {
	console.log(event.type, event.path);
}

import { parseSSEStream } from "@cloudflare/sandbox";
import type { FileWatchSSEEvent } from "@cloudflare/sandbox";

const stream = await sandbox.watch("/workspace/src", { recursive: true });

for await (const event of parseSSEStream<FileWatchSSEEvent>(stream)) {
	console.log(event.type, event.path);
}

Each event includes a type field (create, modify, delete, or move) and the affected path. Move events also include a from field with the original path.

Options

Option	Type	Description
`recursive`	`boolean`	Watch subdirectories. Defaults to `false`.
`include`	`string[]`	Glob patterns to filter events. Omit to receive all events.

Upgrade

To update to the latest version:

npm i @cloudflare/sandbox@latest

For full API details, refer to the Sandbox file watching reference.

Agents, Workers - Agents SDK v0.7.0: Observability rewrite, keepAlive, and waitForMcpConnections

Mon, 02 Mar 2026 00:00:00 GMT

The latest release of the Agents SDK ↗ rewrites observability from scratch with diagnostics_channel, adds keepAlive() to prevent Durable Object eviction during long-running work, and introduces waitForMcpConnections so MCP tools are always available when onChatMessage runs.

Observability rewrite

The previous observability system used console.log() with a custom Observability.emit() interface. v0.7.0 replaces it with structured events published to diagnostics channels — silent by default, zero overhead when nobody is listening.

Every event has a type, payload, and timestamp. Events are routed to seven named channels:

Channel	Event types
`agents:state`	`state:update`
`agents:rpc`	`rpc`, `rpc:error`
`agents:message`	`message:request`, `message:response`, `message:clear`, `message:cancel`, `message:error`, `tool:result`, `tool:approval`
`agents:schedule`	`schedule:create`, `schedule:execute`, `schedule:cancel`, `schedule:retry`, `schedule:error`, `queue:retry`, `queue:error`
`agents:lifecycle`	`connect`, `destroy`
`agents:workflow`	`workflow:start`, `workflow:event`, `workflow:approved`, `workflow:rejected`, `workflow:terminated`, `workflow:paused`, `workflow:resumed`, `workflow:restarted`
`agents:mcp`	`mcp:client:preconnect`, `mcp:client:connect`, `mcp:client:authorize`, `mcp:client:discover`

Use the typed subscribe() helper from agents/observability for type-safe access:

import { subscribe } from "agents/observability";

const unsub = subscribe("rpc", (event) => {
	if (event.type === "rpc") {
		console.log(`RPC call: ${event.payload.method}`);
	}
	if (event.type === "rpc:error") {
		console.error(
			`RPC failed: ${event.payload.method} — ${event.payload.error}`,
		);
	}
});

// Clean up when done
unsub();

import { subscribe } from "agents/observability";

const unsub = subscribe("rpc", (event) => {
	if (event.type === "rpc") {
		console.log(`RPC call: ${event.payload.method}`);
	}
	if (event.type === "rpc:error") {
		console.error(
			`RPC failed: ${event.payload.method} — ${event.payload.error}`,
		);
	}
});

// Clean up when done
unsub();

In production, all diagnostics channel messages are automatically forwarded to Tail Workers — no subscription code needed in the agent itself:

export default {
	async tail(events) {
		for (const event of events) {
			for (const msg of event.diagnosticsChannelEvents) {
				// msg.channel is "agents:rpc", "agents:workflow", etc.
				console.log(msg.timestamp, msg.channel, msg.message);
			}
		}
	},
};

export default {
	async tail(events) {
		for (const event of events) {
			for (const msg of event.diagnosticsChannelEvents) {
				// msg.channel is "agents:rpc", "agents:workflow", etc.
				console.log(msg.timestamp, msg.channel, msg.message);
			}
		}
	},
};

The custom Observability override interface is still supported for users who need to filter or forward events to external services.

For the full event reference, refer to the Diagnostics channels documentation.

`keepAlive()` and `keepAliveWhile()`

Durable Objects are evicted after a period of inactivity (typically 70-140 seconds with no incoming requests, WebSocket messages, or alarms). During long-running operations — streaming LLM responses, waiting on external APIs, running multi-step computations — the agent can be evicted mid-flight.

keepAlive() prevents this by creating a 30-second heartbeat schedule. The alarm firing resets the inactivity timer. Returns a disposer function that cancels the heartbeat when called.

const dispose = await this.keepAlive();
try {
	const result = await longRunningComputation();
	await sendResults(result);
} finally {
	dispose();
}

const dispose = await this.keepAlive();
try {
	const result = await longRunningComputation();
	await sendResults(result);
} finally {
	dispose();
}

keepAliveWhile() wraps an async function with automatic cleanup — the heartbeat starts before the function runs and stops when it completes:

const result = await this.keepAliveWhile(async () => {
	const data = await longRunningComputation();
	return data;
});

const result = await this.keepAliveWhile(async () => {
	const data = await longRunningComputation();
	return data;
});

Key details:

Multiple concurrent callers — Each keepAlive() call returns an independent disposer. Disposing one does not affect others.
AIChatAgent built-in — AIChatAgent automatically calls keepAlive() during streaming responses. You do not need to add it yourself.
Uses the scheduling system — The heartbeat does not conflict with your own schedules. It shows up in getSchedules() if you need to inspect it.

For the full API reference and when-to-use guidance, refer to Schedule tasks — Keeping the agent alive.

`waitForMcpConnections`

AIChatAgent now waits for MCP server connections to settle before calling onChatMessage. This ensures this.mcp.getAITools() returns the full set of tools, especially after Durable Object hibernation when connections are being restored in the background.

export class ChatAgent extends AIChatAgent {
	// Default — waits up to 10 seconds
	// waitForMcpConnections = { timeout: 10_000 };

	// Wait forever
	waitForMcpConnections = true;

	// Disable waiting
	waitForMcpConnections = false;
}

export class ChatAgent extends AIChatAgent {
	// Default — waits up to 10 seconds
	// waitForMcpConnections = { timeout: 10_000 };

	// Wait forever
	waitForMcpConnections = true;

	// Disable waiting
	waitForMcpConnections = false;
}

Value	Behavior
`{ timeout: 10_000 }`	Wait up to 10 seconds (default)
`{ timeout: N }`	Wait up to `N` milliseconds
`true`	Wait indefinitely until all connections ready
`false`	Do not wait (old behavior before 0.2.0)

For lower-level control, call this.mcp.waitForConnections() directly inside onChatMessage instead.

Other improvements

MCP deduplication by name and URL — addMcpServer with HTTP transport now deduplicates on both server name and URL. Calling it with the same name but a different URL creates a new connection. URLs are normalized before comparison (trailing slashes, default ports, hostname case).
callbackHost optional for non-OAuth servers — addMcpServer no longer requires callbackHost when connecting to MCP servers that do not use OAuth.
MCP URL security — Server URLs are validated before connection to prevent SSRF. Private IP ranges, loopback addresses, link-local addresses, and cloud metadata endpoints are blocked.
Custom denial messages — addToolOutput now supports state: "output-error" with errorText for custom denial messages in human-in-the-loop tool approval flows.
requestId in chat options — onChatMessage options now include a requestId for logging and correlating events.

Upgrade

To update to the latest version:

npm i agents@latest @cloudflare/ai-chat@latest

Agents, Workers - Agents SDK v0.6.0: RPC transport for MCP, optional OAuth, hardened schema conversion, and @cloudflare/ai-chat fixes

Wed, 25 Feb 2026 00:00:00 GMT

The latest release of the Agents SDK ↗ lets you define an Agent and an McpAgent in the same Worker and connect them over RPC — no HTTP, no network overhead. It also makes OAuth opt-in for simple MCP connections, hardens the schema converter for production workloads, and ships a batch of @cloudflare/ai-chat reliability fixes.

RPC transport for MCP

You can now connect an Agent to an McpAgent in the same Worker using a Durable Object binding instead of an HTTP URL. The connection stays entirely within the Cloudflare runtime — no network round-trips, no serialization overhead.

Pass the Durable Object namespace directly to addMcpServer:

import { Agent } from "agents";

export class MyAgent extends Agent {
	async onStart() {
		// Connect via DO binding — no HTTP, no network overhead
		await this.addMcpServer("counter", env.MY_MCP);

		// With props for per-user context
		await this.addMcpServer("counter", env.MY_MCP, {
			props: { userId: "user-123", role: "admin" },
		});
	}
}

import { Agent } from "agents";

export class MyAgent extends Agent {
	async onStart() {
		// Connect via DO binding — no HTTP, no network overhead
		await this.addMcpServer("counter", env.MY_MCP);

		// With props for per-user context
		await this.addMcpServer("counter", env.MY_MCP, {
			props: { userId: "user-123", role: "admin" },
		});
	}
}

The addMcpServer method now accepts string | DurableObjectNamespace as the second parameter with full TypeScript overloads, so HTTP and RPC paths are type-safe and cannot be mixed.

Key capabilities:

Hibernation support — RPC connections survive Durable Object hibernation automatically. The binding name and props are persisted to storage and restored on wake-up, matching the behavior of HTTP MCP connections.
Deduplication — Calling addMcpServer with the same server name returns the existing connection instead of creating duplicates. Connection IDs are stable across hibernation restore.
Smaller surface area — The RPC transport internals have been rewritten and reduced from 609 lines to 245 lines. RPCServerTransport now uses JSONRPCMessageSchema from the MCP SDK for validation instead of hand-written checks.

Optional OAuth for MCP connections

addMcpServer() no longer eagerly creates an OAuth provider for every connection. For servers that do not require authentication, a simple call is all you need:

// No callbackHost, no OAuth config — just works
await this.addMcpServer("my-server", "https://mcp.example.com");

// No callbackHost, no OAuth config — just works
await this.addMcpServer("my-server", "https://mcp.example.com");

If the server responds with a 401, the SDK throws a clear error: "This MCP server requires OAuth authentication. Provide callbackHost in addMcpServer options to enable the OAuth flow." The restore-from-storage flow also handles missing callback URLs gracefully, skipping auth provider creation for non-OAuth servers.

Hardened JSON Schema to TypeScript converter

The schema converter used by generateTypes() and getAITools() now handles edge cases that previously caused crashes in production:

Depth and circular reference guards — Prevents stack overflows on recursive or deeply nested schemas
$ref resolution — Supports internal JSON Pointers (#/definitions/..., #/$defs/..., #)
Tuple support — prefixItems (JSON Schema 2020-12) and array items (draft-07)
OpenAPI 3.0 nullable: true — Supported across all schema branches
Per-tool error isolation — One malformed schema cannot crash the full pipeline in generateTypes() or getAITools()
Missing inputSchema fallback — getAITools() falls back to { type: "object" } instead of throwing

`@cloudflare/ai-chat` fixes

Tool denial flow — Denied tool approvals (approved: false) now transition to output-denied with a tool_result, fixing Anthropic provider compatibility. Custom denial messages are supported via state: "output-error" and errorText.
Abort/cancel support — Streaming responses now properly cancel the reader loop when the abort signal fires and send a done signal to the client.
Duplicate message persistence — persistMessages() now reconciles assistant messages by content and order, preventing duplicate rows when clients resend full history.
requestId in OnChatMessageOptions — Handlers can now send properly-tagged error responses for pre-stream failures.
redacted_thinking preservation — The message sanitizer no longer strips Anthropic redacted_thinking blocks.
/get-messages reliability — Endpoint handling moved from a prototype onRequest() override to a constructor wrapper, so it works even when users override onRequest without calling super.onRequest().
Client tool APIs undeprecated — createToolsFromClientSchemas, clientTools, AITool, extractClientToolSchemas, and the tools option on useAgentChat are restored for SDK use cases where tools are defined dynamically at runtime.
jsonSchema initialization — Fixed jsonSchema not initialized error when calling getAITools() in onChatMessage.

Upgrade

To update to the latest version:

npm i agents@latest @cloudflare/ai-chat@latest

Agents, R2, Containers - Backup and restore API for Sandbox SDK

Mon, 23 Feb 2026 00:00:00 GMT

Sandboxes now support createBackup() and restoreBackup() methods for creating and restoring point-in-time snapshots of directories.

This allows you to restore environments quickly. For instance, in order to develop in a sandbox, you may need to include a user's codebase and run a build step. Unfortunately git clone and npm install can take minutes, and you don't want to run these steps every time the user starts their sandbox.

Now, after the initial setup, you can just call createBackup(), then restoreBackup() the next time this environment is needed. This makes it practical to pick up exactly where a user left off, even after days of inactivity, without repeating expensive setup steps.

const sandbox = getSandbox(env.Sandbox, "my-sandbox");

// Make non-trivial changes to the file system
await sandbox.gitCheckout(endUserRepo, { targetDir: "/workspace" });
await sandbox.exec("npm install", { cwd: "/workspace" });

// Create a point-in-time backup of the directory
const backup = await sandbox.createBackup({ dir: "/workspace" });

// Store the handle for later use
await env.KV.put(`backup:${userId}`, JSON.stringify(backup));

// ... in a future session...

// Restore instead of re-cloning and reinstalling
await sandbox.restoreBackup(backup);

Backups are stored in R2 and can take advantage of R2 object lifecycle rules to ensure they do not persist forever.

Key capabilities:

Persist and reuse across sandbox sessions — Easily store backup handles in KV, D1, or Durable Object storage for use in subsequent sessions
Usable across multiple instances — Fork a backup across many sandboxes for parallel work
Named backups — Provide optional human-readable labels for easier management
TTLs — Set time-to-live durations so backups are automatically removed from storage once they are no longer needed

To get started, refer to the backup and restore guide for setup instructions and usage patterns, or the Backups API reference for full method documentation.

Agents, Workers - @cloudflare/codemode v0.1.0: a new runtime agnostic modular architecture

Fri, 20 Feb 2026 00:00:00 GMT

The @cloudflare/codemode ↗ package has been rewritten into a modular, runtime-agnostic SDK.

Code Mode ↗ enables LLMs to write and execute code that orchestrates your tools, instead of calling them one at a time. This can (and does) yield significant token savings, reduces context window pressure and improves overall model performance on a task.

The new Executor interface is runtime agnostic and comes with a prebuilt DynamicWorkerExecutor to run generated code in a Dynamic Worker Loader.

Breaking changes

Removed experimental_codemode() and CodeModeProxy — the package no longer owns an LLM call or model choice
New import path: createCodeTool() is now exported from @cloudflare/codemode/ai

New features

createCodeTool() — Returns a standard AI SDK Tool to use in your AI agents.
Executor interface — Minimal execute(code, fns) contract. Implement for any code sandboxing primitive or runtime.

`DynamicWorkerExecutor`

Runs code in a Dynamic Worker. It comes with the following features:

Network isolation — fetch() and connect() blocked by default (globalOutbound: null) when using DynamicWorkerExecutor
Console capture — console.log/warn/error captured and returned in ExecuteResult.logs
Execution timeout — Configurable via timeout option (default 30s)

Usage

import { createCodeTool } from "@cloudflare/codemode/ai";
import { DynamicWorkerExecutor } from "@cloudflare/codemode";
import { streamText } from "ai";

const executor = new DynamicWorkerExecutor({ loader: env.LOADER });
const codemode = createCodeTool({ tools: myTools, executor });

const result = streamText({
	model,
	tools: { codemode },
	messages,
});

import { createCodeTool } from "@cloudflare/codemode/ai";
import { DynamicWorkerExecutor } from "@cloudflare/codemode";
import { streamText } from "ai";

const executor = new DynamicWorkerExecutor({ loader: env.LOADER });
const codemode = createCodeTool({ tools: myTools, executor });

const result = streamText({
	model,
	tools: { codemode },
	messages,
});

Wrangler configuration

{
	"worker_loaders": [{ "binding": "LOADER" }],
}

[[worker_loaders]]
binding = "LOADER"

See the Code Mode documentation for full API reference and examples.

Upgrade

npm i @cloudflare/codemode@latest

Agents, Workers - Agents SDK v0.5.0: Protocol message control, retry utilities, data parts, and @cloudflare/ai-chat v0.1.0

Tue, 17 Feb 2026 00:00:00 GMT

The latest release of the Agents SDK ↗ adds built-in retry utilities, per-connection protocol message control, and a fully rewritten @cloudflare/ai-chat with data parts, tool approval persistence, and zero breaking changes.

Retry utilities

A new this.retry() method lets you retry any async operation with exponential backoff and jitter. You can pass an optional shouldRetry predicate to bail early on non-retryable errors.

class MyAgent extends Agent {
	async onRequest(request) {
		const data = await this.retry(() => callUnreliableService(), {
			maxAttempts: 4,
			shouldRetry: (err) => !(err instanceof PermanentError),
		});
		return Response.json(data);
	}
}

class MyAgent extends Agent {
	async onRequest(request: Request) {
		const data = await this.retry(() => callUnreliableService(), {
			maxAttempts: 4,
			shouldRetry: (err) => !(err instanceof PermanentError),
		});
		return Response.json(data);
	}
}

Retry options are also available per-task on queue(), schedule(), scheduleEvery(), and addMcpServer():

// Per-task retry configuration, persisted in SQLite alongside the task
await this.schedule(
	Date.now() + 60_000,
	"sendReport",
	{ userId: "abc" },
	{
		retry: { maxAttempts: 5 },
	},
);

// Class-level retry defaults
class MyAgent extends Agent {
	static options = {
		retry: { maxAttempts: 3 },
	};
}

// Per-task retry configuration, persisted in SQLite alongside the task
await this.schedule(Date.now() + 60_000, "sendReport", { userId: "abc" }, {
	retry: { maxAttempts: 5 },
});

// Class-level retry defaults
class MyAgent extends Agent {
	static options = {
		retry: { maxAttempts: 3 },
	};
}

Retry options are validated eagerly at enqueue/schedule time, and invalid values throw immediately. Internal retries have also been added for workflow operations (terminateWorkflow, pauseWorkflow, and others) with Durable Object-aware error detection.

Per-connection protocol message control

Agents automatically send JSON text frames (identity, state, MCP server lists) to every WebSocket connection. You can now suppress these per-connection for clients that cannot handle them — binary-only devices, MQTT clients, or lightweight embedded systems.

class MyAgent extends Agent {
	shouldSendProtocolMessages(connection, ctx) {
		// Suppress protocol messages for MQTT clients
		const subprotocol = ctx.request.headers.get("Sec-WebSocket-Protocol");
		return subprotocol !== "mqtt";
	}
}

class MyAgent extends Agent {
	shouldSendProtocolMessages(connection: Connection, ctx: ConnectionContext) {
		// Suppress protocol messages for MQTT clients
		const subprotocol = ctx.request.headers.get("Sec-WebSocket-Protocol");
		return subprotocol !== "mqtt";
	}
}

Connections with protocol messages disabled still fully participate in RPC and regular messaging. Use isConnectionProtocolEnabled(connection) to check a connection's status at any time. The flag persists across Durable Object hibernation.

See Protocol messages for full documentation.

`@cloudflare/ai-chat` v0.1.0

The first stable release of @cloudflare/ai-chat ships alongside this release with a major refactor of AIChatAgent internals — new ResumableStream class, WebSocket ChatTransport, and simplified SSE parsing — with zero breaking changes. Existing code using AIChatAgent and useAgentChat works as-is.

Key new features:

Data parts — Attach typed JSON blobs (data-*) to messages alongside text. Supports reconciliation (type+id updates in-place), append, and transient parts (ephemeral via onData callback). See Data parts.
Tool approval persistence — The needsApproval approval UI now survives page refresh and DO hibernation. The streaming message is persisted to SQLite when a tool enters approval-requested state.
maxPersistedMessages — Cap SQLite message storage with automatic oldest-message deletion.
body option on useAgentChat — Send custom data with every request (static or dynamic).
Incremental persistence — Hash-based cache to skip redundant SQL writes.
Row size guard — Automatic two-pass compaction when messages approach the SQLite 2 MB limit.
autoContinueAfterToolResult defaults to true — Client-side tool results and tool approvals now automatically trigger a server continuation, matching server-executed tool behavior. Set autoContinueAfterToolResult: false in useAgentChat to restore the previous behavior.

Notable bug fixes:

Resolved stream resumption race conditions
Resolved an issue where setMessages functional updater sent empty arrays
Resolved an issue where client tool schemas were lost after DO hibernation
Resolved InvalidPromptError after tool approval (approval.id was dropped)
Resolved an issue where message metadata was not propagated on broadcast/resume paths
Resolved an issue where clearAll() did not clear in-memory chunk buffers
Resolved an issue where reasoning-delta silently dropped data when reasoning-start was missed during stream resumption

Synchronous queue and schedule getters

getQueue(), getQueues(), getSchedule(), dequeue(), dequeueAll(), and dequeueAllByCallback() were unnecessarily async despite only performing synchronous SQL operations. They now return values directly instead of wrapping them in Promises. This is backward compatible — existing code using await on these methods will continue to work.

Other improvements

Fix TypeScript "excessively deep" error — A depth counter on CanSerialize and IsSerializableParam types bails out to true after 10 levels of recursion, preventing the "Type instantiation is excessively deep" error with deeply nested types like AI SDK CoreMessage[].
POST SSE keepalive — The POST SSE handler now sends event: ping every 30 seconds to keep the connection alive, matching the existing GET SSE handler behavior. This prevents POST response streams from being silently dropped by proxies during long-running tool calls.
Widened peer dependency ranges — Peer dependency ranges across packages have been widened to prevent cascading major bumps during 0.x minor releases. @cloudflare/ai-chat and @cloudflare/codemode are now marked as optional peer dependencies.

Upgrade

To update to the latest version:

npm i agents@latest @cloudflare/ai-chat@latest

Workers, Agents, Workers AI - Introducing GLM-4.7-Flash on Workers AI, @cloudflare/tanstack-ai, and workers-ai-provider v3.1.1

Fri, 13 Feb 2026 00:00:00 GMT

We're excited to announce GLM-4.7-Flash on Workers AI, a fast and efficient text generation model optimized for multilingual dialogue and instruction-following tasks, along with the brand-new @cloudflare/tanstack-ai ↗ package and workers-ai-provider v3.1.1 ↗.

You can now run AI agents entirely on Cloudflare. With GLM-4.7-Flash's multi-turn tool calling support, plus full compatibility with TanStack AI and the Vercel AI SDK, you have everything you need to build agentic applications that run completely at the edge.

GLM-4.7-Flash — Multilingual Text Generation Model

@cf/zai-org/glm-4.7-flash is a multilingual model with a 131,072 token context window, making it ideal for long-form content generation, complex reasoning tasks, and multilingual applications.

Key Features and Use Cases:

Multi-turn Tool Calling for Agents: Build AI agents that can call functions and tools across multiple conversation turns
Multilingual Support: Built to handle content generation in multiple languages effectively
Large Context Window: 131,072 tokens for long-form writing, complex reasoning, and processing long documents
Fast Inference: Optimized for low-latency responses in chatbots and virtual assistants
Instruction Following: Excellent at following complex instructions for code generation and structured tasks

Use GLM-4.7-Flash through the Workers AI binding (env.AI.run()), the REST API at /run or /v1/chat/completions, AI Gateway, or via workers-ai-provider for the Vercel AI SDK.

Pricing is available on the model page or pricing page.

@cloudflare/tanstack-ai v0.1.1 — TanStack AI adapters for Workers AI and AI Gateway

We've released @cloudflare/tanstack-ai, a new package that brings Workers AI and AI Gateway support to TanStack AI ↗. This provides a framework-agnostic alternative for developers who prefer TanStack's approach to building AI applications.

Workers AI adapters support four configuration modes — plain binding (env.AI), plain REST, AI Gateway binding (env.AI.gateway(id)), and AI Gateway REST — across all capabilities:

Chat (createWorkersAiChat) — Streaming chat completions with tool calling, structured output, and reasoning text streaming.
Image generation (createWorkersAiImage) — Text-to-image models.
Transcription (createWorkersAiTranscription) — Speech-to-text.
Text-to-speech (createWorkersAiTts) — Audio generation.
Summarization (createWorkersAiSummarize) — Text summarization.

AI Gateway adapters route requests from third-party providers — OpenAI, Anthropic, Gemini, Grok, and OpenRouter — through Cloudflare AI Gateway for caching, rate limiting, and unified billing.

To get started:

npm install @cloudflare/tanstack-ai @tanstack/ai

workers-ai-provider v3.1.1 — transcription, speech, reranking, and reliability

The Workers AI provider for the Vercel AI SDK ↗ now supports three new capabilities beyond chat and image generation:

Transcription (provider.transcription(model)) — Speech-to-text with automatic handling of model-specific input formats across binding and REST paths.
Text-to-speech (provider.speech(model)) — Audio generation with support for voice and speed options.
Reranking (provider.reranking(model)) — Document reranking for RAG pipelines and search result ordering.

import { createWorkersAI } from "workers-ai-provider";
import {
	experimental_transcribe,
	experimental_generateSpeech,
	rerank,
} from "ai";

const workersai = createWorkersAI({ binding: env.AI });

const transcript = await experimental_transcribe({
	model: workersai.transcription("@cf/openai/whisper-large-v3-turbo"),
	audio: audioData,
	mediaType: "audio/wav",
});

const speech = await experimental_generateSpeech({
	model: workersai.speech("@cf/deepgram/aura-1"),
	text: "Hello world",
	voice: "asteria",
});

const ranked = await rerank({
	model: workersai.reranking("@cf/baai/bge-reranker-base"),
	query: "What is machine learning?",
	documents: ["ML is a branch of AI.", "The weather is sunny."],
});

This release also includes a comprehensive reliability overhaul (v3.0.5):

Fixed streaming — Responses now stream token-by-token instead of buffering all chunks, using a proper TransformStream pipeline with backpressure.
Fixed tool calling — Resolved issues with tool call ID sanitization, conversation history preservation, and a heuristic that silently fell back to non-streaming mode when tools were defined.
Premature stream termination detection — Streams that end unexpectedly now report finishReason: "error" instead of silently reporting "stop".
AI Search support — Added createAISearch as the canonical export (renamed from AutoRAG). createAutoRAG still works with a deprecation warning.

To upgrade:

npm install workers-ai-provider@latest ai

Resources

Agents, Workers - Agents SDK v0.4.0: Readonly connections, MCP security improvements, x402 v2 migration, and custom MCP OAuth providers

Mon, 09 Feb 2026 00:00:00 GMT

The latest release of the Agents SDK ↗ brings readonly connections, MCP protocol and security improvements, x402 payment protocol v2 migration, and the ability to customize OAuth for MCP server connections.

Readonly connections

Agents can now restrict WebSocket clients to read-only access, preventing them from modifying agent state. This is useful for dashboards, spectator views, or any scenario where clients should observe but not mutate.

New hooks: shouldConnectionBeReadonly, setConnectionReadonly, isConnectionReadonly. Readonly connections block both client-side setState() and mutating @callable() methods, and the readonly flag survives hibernation.

class MyAgent extends Agent {
	shouldConnectionBeReadonly(connection) {
		// Make spectators readonly
		return connection.url.includes("spectator");
	}
}

class MyAgent extends Agent {
	shouldConnectionBeReadonly(connection) {
		// Make spectators readonly
		return connection.url.includes("spectator");
	}
}

Custom MCP OAuth providers

The new createMcpOAuthProvider method on the Agent class allows subclasses to override the default OAuth provider used when connecting to MCP servers. This enables custom authentication strategies such as pre-registered client credentials or mTLS, beyond the built-in dynamic client registration.

class MyAgent extends Agent {
	createMcpOAuthProvider(callbackUrl) {
		return new MyCustomOAuthProvider(this.ctx.storage, this.name, callbackUrl);
	}
}

class MyAgent extends Agent {
	createMcpOAuthProvider(callbackUrl: string): AgentMcpOAuthProvider {
		return new MyCustomOAuthProvider(this.ctx.storage, this.name, callbackUrl);
	}
}

MCP SDK upgrade to 1.26.0

Upgraded the MCP SDK to 1.26.0 to prevent cross-client response leakage. Stateless MCP Servers should now create a new McpServer instance per request instead of sharing a single instance. A guard is added in this version of the MCP SDK which will prevent connection to a Server instance that has already been connected to a transport. Developers will need to modify their code if they declare their McpServer instance as a global variable.

MCP OAuth callback URL security fix

Added callbackPath option to addMcpServer to prevent instance name leakage in MCP OAuth callback URLs. When sendIdentityOnConnect is false, callbackPath is now required — the default callback URL would expose the instance name, undermining the security intent. Also fixes callback request detection to match via the state parameter instead of a loose /callback URL substring check, enabling custom callback paths.

Deprecate `onStateUpdate` in favor of `onStateChanged`

onStateChanged is a drop-in rename of onStateUpdate (same signature, same behavior). onStateUpdate still works but emits a one-time console warning per class. validateStateChange rejections now propagate a CF_AGENT_STATE_ERROR message back to the client.

x402 v2 migration

Migrated the x402 MCP payment integration from the legacy x402 package to @x402/core and @x402/evm v2.

Breaking changes for x402 users:

Peer dependencies changed: replace x402 with @x402/core and @x402/evm
PaymentRequirements type now uses v2 fields (e.g. amount instead of maxAmountRequired)
X402ClientConfig.account type changed from viem.Account to ClientEvmSigner (structurally compatible with privateKeyToAccount())

npm uninstall x402
npm install @x402/core @x402/evm

Network identifiers now accept both legacy names and CAIP-2 format:

// Legacy name (auto-converted)
{
	network: "base-sepolia",
}

// CAIP-2 format (preferred)
{
	network: "eip155:84532",
}

Other x402 changes:

X402ClientConfig.network is now optional — the client auto-selects from available payment requirements
Server-side lazy initialization: facilitator connection is deferred until the first paid tool invocation
Payment tokens support both v2 (PAYMENT-SIGNATURE) and v1 (X-PAYMENT) HTTP headers
Added normalizeNetwork export for converting legacy network names to CAIP-2 format
Re-exports PaymentRequirements, PaymentRequired, Network, FacilitatorConfig, and ClientEvmSigner from agents/x402

Other improvements

Fix useAgent and AgentClient crashing when using basePath routing
CORS handling delegated to partyserver's native support (simpler, more reliable)
Client-side onStateUpdateError callback for handling rejected state updates

Upgrade

To update to the latest version:

npm i agents@latest

Agents - Interactive browser terminals in Sandboxes

Mon, 09 Feb 2026 00:00:00 GMT

The Sandbox SDK ↗ now supports PTY (pseudo-terminal) passthrough, enabling browser-based terminal UIs to connect to sandbox shells via WebSocket.

`sandbox.terminal(request)`

The new terminal() method proxies a WebSocket upgrade to the container's PTY endpoint, with output buffering for replay on reconnect.

// Worker: proxy WebSocket to container terminal
return sandbox.terminal(request, { cols: 80, rows: 24 });

// Worker: proxy WebSocket to container terminal
return sandbox.terminal(request, { cols: 80, rows: 24 });

Multiple terminals per sandbox

Each session can have its own terminal with an isolated working directory and environment, so users can run separate shells side-by-side in the same container.

// Multiple isolated terminals in the same sandbox
const dev = await sandbox.getSession("dev");
return dev.terminal(request);

// Multiple isolated terminals in the same sandbox
const dev = await sandbox.getSession("dev");
return dev.terminal(request);

xterm.js addon

The new @cloudflare/sandbox/xterm export provides a SandboxAddon for xterm.js ↗ with automatic reconnection (exponential backoff + jitter), buffered output replay, and resize forwarding.

import { SandboxAddon } from "@cloudflare/sandbox/xterm";

const addon = new SandboxAddon({
	getWebSocketUrl: ({ sandboxId, origin }) =>
		`${origin}/ws/terminal?id=${sandboxId}`,
	onStateChange: (state, error) => updateUI(state),
});
terminal.loadAddon(addon);
addon.connect({ sandboxId: "my-sandbox" });

import { SandboxAddon } from "@cloudflare/sandbox/xterm";

const addon = new SandboxAddon({
	getWebSocketUrl: ({ sandboxId, origin }) =>
		`${origin}/ws/terminal?id=${sandboxId}`,
	onStateChange: (state, error) => updateUI(state),
});
terminal.loadAddon(addon);
addon.connect({ sandboxId: "my-sandbox" });

Upgrade

To update to the latest version:

npm i @cloudflare/sandbox@latest

Agents, Workflows - Agents SDK v0.3.7: Workflows integration, synchronous state, and scheduleEvery()

Tue, 03 Feb 2026 00:00:00 GMT

The latest release of the Agents SDK ↗ brings first-class support for Cloudflare Workflows, synchronous state management, and new scheduling capabilities.

Cloudflare Workflows integration

Agents excel at real-time communication and state management. Workflows excel at durable execution. Together, they enable powerful patterns where Agents handle WebSocket connections while Workflows handle long-running tasks, retries, and human-in-the-loop flows.

Use the new AgentWorkflow class to define workflows with typed access to your Agent:

import { AgentWorkflow } from "agents/workflows";

export class ProcessingWorkflow extends AgentWorkflow {
	async run(event, step) {
		// Call Agent methods via RPC
		await this.agent.updateStatus(event.payload.taskId, "processing");

		// Non-durable: progress reporting to clients
		await this.reportProgress({ step: "process", percent: 0.5 });
		this.broadcastToClients({ type: "update", taskId: event.payload.taskId });

		// Durable via step: idempotent, won't repeat on retry
		await step.mergeAgentState({ taskProgress: 0.5 });

		const result = await step.do("process", async () => {
			return processData(event.payload.data);
		});

		await step.reportComplete(result);
		return result;
	}
}

import { AgentWorkflow } from "agents/workflows";
import type { AgentWorkflowEvent, AgentWorkflowStep } from "agents/workflows";

export class ProcessingWorkflow extends AgentWorkflow<MyAgent, TaskParams> {
	async run(event: AgentWorkflowEvent<TaskParams>, step: AgentWorkflowStep) {
		// Call Agent methods via RPC
		await this.agent.updateStatus(event.payload.taskId, "processing");

		// Non-durable: progress reporting to clients
		await this.reportProgress({ step: "process", percent: 0.5 });
		this.broadcastToClients({ type: "update", taskId: event.payload.taskId });

		// Durable via step: idempotent, won't repeat on retry
		await step.mergeAgentState({ taskProgress: 0.5 });

		const result = await step.do("process", async () => {
			return processData(event.payload.data);
		});

		await step.reportComplete(result);
		return result;
	}
}

Start workflows from your Agent with runWorkflow() and handle lifecycle events:

export class MyAgent extends Agent {
	async startTask(taskId, data) {
		const instanceId = await this.runWorkflow("PROCESSING_WORKFLOW", {
			taskId,
			data,
		});
		return { instanceId };
	}

	async onWorkflowProgress(workflowName, instanceId, progress) {
		this.broadcast(JSON.stringify({ type: "progress", progress }));
	}

	async onWorkflowComplete(workflowName, instanceId, result) {
		console.log(`Workflow ${instanceId} completed`);
	}

	async onWorkflowError(workflowName, instanceId, error) {
		console.error(`Workflow ${instanceId} failed:`, error);
	}
}

export class MyAgent extends Agent {
	async startTask(taskId: string, data: string) {
		const instanceId = await this.runWorkflow("PROCESSING_WORKFLOW", {
			taskId,
			data,
		});
		return { instanceId };
	}

	async onWorkflowProgress(
		workflowName: string,
		instanceId: string,
		progress: unknown,
	) {
		this.broadcast(JSON.stringify({ type: "progress", progress }));
	}

	async onWorkflowComplete(
		workflowName: string,
		instanceId: string,
		result?: unknown,
	) {
		console.log(`Workflow ${instanceId} completed`);
	}

	async onWorkflowError(
		workflowName: string,
		instanceId: string,
		error: unknown,
	) {
		console.error(`Workflow ${instanceId} failed:`, error);
	}
}

Key workflow methods on your Agent:

runWorkflow(workflowName, params, options?) — Start a workflow with optional metadata
getWorkflow(workflowId) / getWorkflows(criteria?) — Query workflows with cursor-based pagination
approveWorkflow(workflowId) / rejectWorkflow(workflowId) — Human-in-the-loop approval flows
pauseWorkflow(), resumeWorkflow(), terminateWorkflow() — Workflow control

Synchronous setState()

State updates are now synchronous with a new validateStateChange() validation hook:

export class MyAgent extends Agent {
	validateStateChange(oldState, newState) {
		// Return false to reject the change
		if (newState.count < 0) return false;
		// Return modified state to transform
		return { ...newState, lastUpdated: Date.now() };
	}
}

export class MyAgent extends Agent<Env, State> {
	validateStateChange(oldState: State, newState: State): State | false {
		// Return false to reject the change
		if (newState.count < 0) return false;
		// Return modified state to transform
		return { ...newState, lastUpdated: Date.now() };
	}
}

scheduleEvery() for recurring tasks

The new scheduleEvery() method enables fixed-interval recurring tasks with built-in overlap prevention:

// Run every 5 minutes
await this.scheduleEvery("syncData", 5 * 60 * 1000, { source: "api" });

// Run every 5 minutes
await this.scheduleEvery("syncData", 5 * 60 * 1000, { source: "api" });

Callable system improvements

Client-side RPC timeout — Set timeouts on callable method invocations
StreamingResponse.error(message) — Graceful stream error signaling
getCallableMethods() — Introspection API for discovering callable methods
Connection close handling — Pending calls are automatically rejected on disconnect

await agent.call("method", [args], {
	timeout: 5000,
	stream: { onChunk, onDone, onError },
});

await agent.call("method", [args], {
	timeout: 5000,
	stream: { onChunk, onDone, onError },
});

Email and routing enhancements

Secure email reply routing — Email replies are now secured with HMAC-SHA256 signed headers, preventing unauthorized routing of emails to agent instances.

Routing improvements:

basePath option to bypass default URL construction for custom routing
Server-sent identity — Agents send name and agent type on connect
New onIdentity and onIdentityChange callbacks on the client

const agent = useAgent({
	basePath: "user",
	onIdentity: (name, agentType) => console.log(`Connected to ${name}`),
});

const agent = useAgent({
	basePath: "user",
	onIdentity: (name, agentType) => console.log(`Connected to ${name}`),
});

Upgrade

To update to the latest version:

npm i agents@latest

For the complete Workflows API reference and patterns, see Run Workflows.

Agents, Workers - Agents SDK v0.3.0, workers-ai-provider v3.0.0, and ai-gateway-provider v3.0.0 with AI SDK v6 support

Mon, 22 Dec 2025 00:00:00 GMT

We've shipped a new release for the Agents SDK ↗ v0.3.0 bringing full compatibility with AI SDK v6 ↗ and introducing the unified tool pattern, dynamic tool approval, and enhanced React hooks with improved tool handling.

This release includes improved streaming and tool support, dynamic tool approval (for "human in the loop" systems), enhanced React hooks with onToolCall callback, improved error handling for streaming responses, and seamless migration from v5 patterns.

This makes it ideal for building production AI chat interfaces with Cloudflare Workers AI models, agent workflows, human-in-the-loop systems, or any application requiring reliable tool execution and approval workflows.

Additionally, we've updated workers-ai-provider v3.0.0, the official provider for Cloudflare Workers AI models, and ai-gateway-provider v3.0.0, the provider for Cloudflare AI Gateway, to be compatible with AI SDK v6.

Agents SDK v0.3.0

Unified Tool Pattern

AI SDK v6 introduces a unified tool pattern where all tools are defined on the server using the tool() function. This replaces the previous client-side AITool pattern.

Server-Side Tool Definition

import { tool } from "ai";
import { z } from "zod";

// Server: Define ALL tools on the server
const tools = {
	// Server-executed tool
	getWeather: tool({
		description: "Get weather for a city",
		inputSchema: z.object({ city: z.string() }),
		execute: async ({ city }) => fetchWeather(city)
	}),

	// Client-executed tool (no execute = client handles via onToolCall)
	getLocation: tool({
		description: "Get user location from browser",
		inputSchema: z.object({})
		// No execute function
	}),

	// Tool requiring approval (dynamic based on input)
	processPayment: tool({
		description: "Process a payment",
		inputSchema: z.object({ amount: z.number() }),
		needsApproval: async ({ amount }) => amount > 100,
		execute: async ({ amount }) => charge(amount)
	})
};

Client-Side Tool Handling

// Client: Handle client-side tools via onToolCall callback
import { useAgentChat } from "agents/ai-react";

const { messages, sendMessage, addToolOutput } = useAgentChat({
	agent,
	onToolCall: async ({ toolCall, addToolOutput }) => {
		if (toolCall.toolName === "getLocation") {
			const position = await new Promise((resolve, reject) => {
				navigator.geolocation.getCurrentPosition(resolve, reject);
			});
			addToolOutput({
				toolCallId: toolCall.toolCallId,
				output: {
					lat: position.coords.latitude,
					lng: position.coords.longitude
				}
			});
		}
	}
});

Key benefits of the unified tool pattern:

Server-defined tools: All tools are defined in one place on the server
Dynamic approval: Use needsApproval to conditionally require user confirmation
Cleaner client code: Use onToolCall callback instead of managing tool configs
Type safety: Full TypeScript support with proper tool typing

useAgentChat(options)

Creates a new chat interface with enhanced v6 capabilities.

// Basic chat setup with onToolCall
const { messages, sendMessage, addToolOutput } = useAgentChat({
	agent,
	onToolCall: async ({ toolCall, addToolOutput }) => {
		// Handle client-side tool execution
		await addToolOutput({
			toolCallId: toolCall.toolCallId,
			output: { result: "success" }
		});
	}
});

Dynamic Tool Approval

Use needsApproval on server tools to conditionally require user confirmation:

const paymentTool = tool({
	description: "Process a payment",
	inputSchema: z.object({
		amount: z.number(),
		recipient: z.string()
	}),
	needsApproval: async ({ amount }) => amount > 1000,
	execute: async ({ amount, recipient }) => {
		return await processPayment(amount, recipient);
	}
});

Tool Confirmation Detection

The isToolUIPart and getToolName functions now check both static and dynamic tool parts:

import { isToolUIPart, getToolName } from "ai";

const pendingToolCallConfirmation = messages.some((m) =>
	m.parts?.some(
		(part) => isToolUIPart(part) && part.state === "input-available",
	),
);

// Handle tool confirmation
if (pendingToolCallConfirmation) {
	await addToolOutput({
		toolCallId: part.toolCallId,
		output: "User approved the action"
	});
}

If you need the v5 behavior (static-only checks), use the new functions:

import { isStaticToolUIPart, getStaticToolName } from "ai";

convertToModelMessages() is now async

The convertToModelMessages() function is now asynchronous. Update all calls to await the result:

import { convertToModelMessages } from "ai";

const result = streamText({
	messages: await convertToModelMessages(this.messages),
	model: openai("gpt-4o")
});

ModelMessage type

The CoreMessage type has been removed. Use ModelMessage instead:

import { convertToModelMessages, type ModelMessage } from "ai";

const modelMessages: ModelMessage[] = await convertToModelMessages(messages);

generateObject mode option removed

The mode option for generateObject has been removed:

// Before (v5)
const result = await generateObject({
	mode: "json",
	model,
	schema,
	prompt
});

// After (v6)
const result = await generateObject({
	model,
	schema,
	prompt
});

Structured Output with generateText

While generateObject and streamObject are still functional, the recommended approach is to use generateText/streamText with the Output.object() helper:

import { generateText, Output, stepCountIs } from "ai";

const { output } = await generateText({
	model: openai("gpt-4"),
	output: Output.object({
		schema: z.object({ name: z.string() })
	}),
	stopWhen: stepCountIs(2),
	prompt: "Generate a name"
});

Note: When using structured output with generateText, you must configure multiple steps with stopWhen because generating the structured output is itself a step.

workers-ai-provider v3.0.0

Seamless integration with Cloudflare Workers AI models through the updated workers-ai-provider v3.0.0 with AI SDK v6 support.

Model Setup with Workers AI

Use Cloudflare Workers AI models directly in your agent workflows:

import { createWorkersAI } from "workers-ai-provider";
import { useAgentChat } from "agents/ai-react";

// Create Workers AI model (v3.0.0 - enhanced v6 internals)
const model = createWorkersAI({
	binding: env.AI,
})("@cf/meta/llama-3.2-3b-instruct");

Enhanced File and Image Support

Workers AI models now support v6 file handling with automatic conversion:

// Send images and files to Workers AI models
sendMessage({
	role: "user",
	parts: [
		{ type: "text", text: "Analyze this image:" },
		{
			type: "file",
			data: imageBuffer,
			mediaType: "image/jpeg",
		},
	],
});

// Workers AI provider automatically converts to proper format

Streaming with Workers AI

Enhanced streaming support with automatic warning detection:

// Streaming with Workers AI models
const result = await streamText({
	model: createWorkersAI({ binding: env.AI })("@cf/meta/llama-3.2-3b-instruct"),
	messages: await convertToModelMessages(messages),
	onChunk: (chunk) => {
		// Enhanced streaming with warning handling
		console.log(chunk);
	},
});

ai-gateway-provider v3.0.0

The ai-gateway-provider v3.0.0 now supports AI SDK v6, enabling you to use Cloudflare AI Gateway with multiple AI providers including Anthropic, Azure, AWS Bedrock, Google Vertex, and Perplexity.

AI Gateway Setup

Use Cloudflare AI Gateway to add analytics, caching, and rate limiting to your AI applications:

import { createAIGateway } from "ai-gateway-provider";

// Create AI Gateway provider (v3.0.0 - enhanced v6 internals)
const model = createAIGateway({
	gatewayUrl: "https://gateway.ai.cloudflare.com/v1/your-account-id/gateway",
	headers: {
		"Authorization": `Bearer ${env.AI_GATEWAY_TOKEN}`
	}
})({
	provider: "openai",
	model: "gpt-4o"
});

Migration from v5

Deprecated APIs

The following APIs are deprecated in favor of the unified tool pattern:

Deprecated	Replacement
`AITool` type	Use AI SDK's `tool()` function on server
`extractClientToolSchemas()`	Define tools on server, no client schemas needed
`createToolsFromClientSchemas()`	Define tools on server with `tool()`
`toolsRequiringConfirmation` option	Use `needsApproval` on server tools
`experimental_automaticToolResolution`	Use `onToolCall` callback
`tools` option in `useAgentChat`	Use `onToolCall` for client-side execution
`addToolResult()`	Use `addToolOutput()`

Breaking Changes Summary

Unified Tool Pattern: All tools must be defined on the server using tool()
convertToModelMessages() is async: Add await to all calls
CoreMessage removed: Use ModelMessage instead
generateObject mode removed: Remove mode option
isToolUIPart behavior changed: Now checks both static and dynamic tool parts

Installation

Update your dependencies to use the latest versions:

npm install agents@^0.3.0 workers-ai-provider@^3.0.0 ai-gateway-provider@^3.0.0 ai@^6.0.0 @ai-sdk/react@^3.0.0 @ai-sdk/openai@^3.0.0

Resources

Migration Guide ↗ - Comprehensive migration documentation from v5 to v6
AI SDK v6 Documentation ↗ - Official AI SDK migration guide
AI SDK v6 Announcement ↗ - Learn about new features in v6
AI SDK Documentation ↗ - Complete AI SDK reference
GitHub Issues ↗ - Report bugs or request features

Feedback Welcome

We'd love your feedback! We're particularly interested in feedback on:

Migration experience - How smooth was the upgrade from v5 to v6?
Unified tool pattern - How does the new server-defined tool pattern work for you?
Dynamic tool approval - Does the needsApproval feature meet your needs?
AI Gateway integration - How well does the new provider work with your setup?

Agents, Workers - Agents SDK v0.2.24 with resumable streaming, MCP improvements, and schedule fixes

Wed, 26 Nov 2025 00:00:00 GMT

The latest release of @cloudflare/agents ↗ brings resumable streaming, significant MCP client improvements, and critical fixes for schedules and Durable Object lifecycle management.

Resumable streaming

AIChatAgent now supports resumable streaming, allowing clients to reconnect and continue receiving streamed responses without losing data. This is useful for:

Long-running AI responses
Users on unreliable networks
Users switching between devices mid-conversation
Background tasks where users navigate away and return
Real-time collaboration where multiple clients need to stay in sync

Streams are maintained across page refreshes, broken connections, and syncing across open tabs and devices.

Other improvements

Default JSON schema validator added to MCP client
Schedules ↗ can now safely destroy the agent

MCP client API improvements

The MCPClientManager API has been redesigned for better clarity and control:

New registerServer() method: Register MCP servers without immediately connecting
New connectToServer() method: Establish connections to registered servers
Improved reconnect logic: restoreConnectionsFromStorage() now properly handles failed connections

// Register a server to Agent
const { id } = await this.mcp.registerServer({
	name: "my-server",
	url: "https://my-mcp-server.example.com",
});

// Connect when ready
await this.mcp.connectToServer(id);

// Discover tools, prompts and resources
await this.mcp.discoverIfConnected(id);

The SDK now includes a formalized MCPConnectionState enum with states: idle, connecting, authenticating, connected, discovering, and ready.

Enhanced MCP discovery

MCP discovery fetches the available tools, prompts, and resources from an MCP server so your agent knows what capabilities are available. The MCPClientConnection class now includes a dedicated discover() method with improved reliability:

Supports cancellation via AbortController
Configurable timeout (default 15s)
Discovery failures now throw errors immediately instead of silently continuing

Bug fixes

Fixed a bug where schedules ↗ meant to fire immediately with this.schedule(0, ...) or this.schedule(new Date(), ...) would not fire
Fixed an issue where schedules that took longer than 30 seconds would occasionally time out
Fixed SSE transport now properly forwards session IDs and request headers
Fixed AI SDK stream events conversion to UIMessageStreamPart

Upgrade

To update to the latest version:

npm i agents@latest

Agents, Workers - Agents SDK v0.1.0 and workers-ai-provider v2.0.0 with AI SDK v5 support

Wed, 10 Sep 2025 00:00:00 GMT

We've shipped a new release for the Agents SDK ↗ bringing full compatibility with AI SDK v5 ↗ and introducing automatic message migration that handles all legacy formats transparently.

This release includes improved streaming and tool support, tool confirmation detection (for "human in the loop" systems), enhanced React hooks with automatic tool resolution, improved error handling for streaming responses, and seamless migration utilities that work behind the scenes.

Additionally, we've updated workers-ai-provider v2.0.0, the official provider for Cloudflare Workers AI models, to be compatible with AI SDK v5.

useAgentChat(options)

Creates a new chat interface with enhanced v5 capabilities.

// Basic chat setup
const { messages, sendMessage, addToolResult } = useAgentChat({
	agent,
	experimental_automaticToolResolution: true,
	tools,
});

// With custom tool confirmation
const chat = useAgentChat({
	agent,
	experimental_automaticToolResolution: true,
	toolsRequiringConfirmation: ["dangerousOperation"],
});

Automatic Tool Resolution

Tools are automatically categorized based on their configuration:

const tools = {
	// Auto-executes (has execute function)
	getLocalTime: {
		description: "Get current local time",
		inputSchema: z.object({}),
		execute: async () => new Date().toLocaleString(),
	},

	// Requires confirmation (no execute function)
	deleteFile: {
		description: "Delete a file from the system",
		inputSchema: z.object({
			filename: z.string(),
		}),
	},

	// Server-executed (no client confirmation)
	analyzeData: {
		description: "Analyze dataset on server",
		inputSchema: z.object({ data: z.array(z.number()) }),
		serverExecuted: true,
	},
} satisfies Record<string, AITool>;

Message Handling

Send messages using the new v5 format with parts array:

// Text message
sendMessage({
	role: "user",
	parts: [{ type: "text", text: "Hello, assistant!" }],
});

// Multi-part message with file
sendMessage({
	role: "user",
	parts: [
		{ type: "text", text: "Analyze this image:" },
		{ type: "image", image: imageData },
	],
});

Tool Confirmation Detection

Simplified logic for detecting pending tool confirmations:

const pendingToolCallConfirmation = messages.some((m) =>
	m.parts?.some(
		(part) => isToolUIPart(part) && part.state === "input-available",
	),
);

// Handle tool confirmation
if (pendingToolCallConfirmation) {
	await addToolResult({
		toolCallId: part.toolCallId,
		tool: getToolName(part),
		output: "User approved the action",
	});
}

Automatic Message Migration

Seamlessly handle legacy message formats without code changes.

// All these formats are automatically converted:

// Legacy v4 string content
const legacyMessage = {
	role: "user",
	content: "Hello world",
};

// Legacy v4 with tool calls
const legacyWithTools = {
	role: "assistant",
	content: "",
	toolInvocations: [
		{
			toolCallId: "123",
			toolName: "weather",
			args: { city: "SF" },
			state: "result",
			result: "Sunny, 72°F",
		},
	],
};

// Automatically becomes v5 format:
// {
//   role: "assistant",
//   parts: [{
//     type: "tool-call",
//     toolCallId: "123",
//     toolName: "weather",
//     args: { city: "SF" },
//     state: "result",
//     result: "Sunny, 72°F"
//   }]
// }

Tool Definition Updates

Migrate tool definitions to use the new inputSchema property.

// Before (AI SDK v4)
const tools = {
	weather: {
		description: "Get weather information",
		parameters: z.object({
			city: z.string(),
		}),
		execute: async (args) => {
			return await getWeather(args.city);
		},
	},
};

// After (AI SDK v5)
const tools = {
	weather: {
		description: "Get weather information",
		inputSchema: z.object({
			city: z.string(),
		}),
		execute: async (args) => {
			return await getWeather(args.city);
		},
	},
};

Cloudflare Workers AI Integration

Seamless integration with Cloudflare Workers AI models through the updated workers-ai-provider v2.0.0.

Model Setup with Workers AI

Use Cloudflare Workers AI models directly in your agent workflows:

import { createWorkersAI } from "workers-ai-provider";
import { useAgentChat } from "agents/ai-react";

// Create Workers AI model (v2.0.0 - same API, enhanced v5 internals)
const model = createWorkersAI({
	binding: env.AI,
})("@cf/meta/llama-3.2-3b-instruct");

Enhanced File and Image Support

Workers AI models now support v5 file handling with automatic conversion:

// Send images and files to Workers AI models
sendMessage({
	role: "user",
	parts: [
		{ type: "text", text: "Analyze this image:" },
		{
			type: "file",
			data: imageBuffer,
			mediaType: "image/jpeg",
		},
	],
});

// Workers AI provider automatically converts to proper format

Streaming with Workers AI

Enhanced streaming support with automatic warning detection:

// Streaming with Workers AI models
const result = await streamText({
	model: createWorkersAI({ binding: env.AI })("@cf/meta/llama-3.2-3b-instruct"),
	messages,
	onChunk: (chunk) => {
		// Enhanced streaming with warning handling
		console.log(chunk);
	},
});

Import Updates

Update your imports to use the new v5 types:

// Before (AI SDK v4)
import type { Message } from "ai";
import { useChat } from "ai/react";

// After (AI SDK v5)
import type { UIMessage } from "ai";
// or alias for compatibility
import type { UIMessage as Message } from "ai";
import { useChat } from "@ai-sdk/react";

Resources

Migration Guide ↗ - Comprehensive migration documentation
AI SDK v5 Documentation ↗ - Official AI SDK migration guide
An Example PR showing the migration from AI SDK v4 to v5 ↗
GitHub Issues ↗ - Report bugs or request features

Feedback Welcome

We'd love your feedback! We're particularly interested in feedback on:

Migration experience - How smooth was the upgrade process?
Tool confirmation workflow - Does the new automatic detection work as expected?
Message format handling - Any edge cases with legacy message conversion?

Agents, Workers - Agents SDK adds MCP Elicitation support, http-streamable support, task queues, email integration and more

Tue, 05 Aug 2025 00:00:00 GMT

The latest releases of @cloudflare/agents ↗ brings major improvements to MCP transport protocols support and agents connectivity. Key updates include:

MCP elicitation support

MCP servers can now request user input during tool execution, enabling interactive workflows like confirmations, forms, and multi-step processes. This feature uses durable storage to preserve elicitation state even during agent hibernation, ensuring seamless user interactions across agent lifecycle events.

// Request user confirmation via elicitation
const confirmation = await this.elicitInput({
	message: `Are you sure you want to increment the counter by ${amount}?`,
	requestedSchema: {
		type: "object",
		properties: {
			confirmed: {
				type: "boolean",
				title: "Confirm increment",
				description: "Check to confirm the increment",
			},
		},
		required: ["confirmed"],
	},
});

Check out our demo ↗ to see elicitation in action.

HTTP streamable transport for MCP

MCP now supports HTTP streamable transport which is recommended over SSE. This transport type offers:

Better performance: More efficient data streaming and reduced overhead
Improved reliability: Enhanced connection stability and error recover- Automatic fallback: If streamable transport is not available, it gracefully falls back to SSE

export default MyMCP.serve("/mcp", {
	binding: "MyMCP",
});

The SDK automatically selects the best available transport method, gracefully falling back from streamable-http to SSE when needed.

Enhanced MCP connectivity

Significant improvements to MCP server connections and transport reliability:

Auto transport selection: Automatically determines the best transport method, falling back from streamable-http to SSE as needed
Improved error handling: Better connection state management and error reporting for MCP servers
Reliable prop updates: Centralized agent property updates ensure consistency across different contexts

Lightweight .queue for fast task deferral

You can use .queue() to enqueue background work — ideal for tasks like processing user messages, sending notifications etc.

class MyAgent extends Agent {
	doSomethingExpensive(payload) {
		// a long running process that you want to run in the background
	}

	queueSomething() {
		await this.queue("doSomethingExpensive", somePayload); // this will NOT block further execution, and runs in the background
		await this.queue("doSomethingExpensive", someOtherPayload); // the callback will NOT run until the previous callback is complete
		// ... call as many times as you want
	}
}

Want to try it yourself? Just define a method like processMessage in your agent, and you’re ready to scale.

New email adapter

Want to build an AI agent that can receive and respond to emails automatically? With the new email adapter and onEmail lifecycle method, now you can.

export class EmailAgent extends Agent {
	async onEmail(email: AgentEmail) {
		const raw = await email.getRaw();
		const parsed = await PostalMime.parse(raw);

		// create a response based on the email contents
		// and then send a reply

		await this.replyToEmail(email, {
			fromName: "Email Agent",
			body: `Thanks for your email! You've sent us "${parsed.subject}". We'll process it shortly.`,
		});
	}
}

You route incoming mail like this:

export default {
	async email(email, env) {
		await routeAgentEmail(email, env, {
			resolver: createAddressBasedEmailResolver("EmailAgent"),
		});
	},
};

You can find a full example here ↗.

Automatic context wrapping for custom methods

Custom methods are now automatically wrapped with the agent's context, so calling getCurrentAgent() should work regardless of where in an agent's lifecycle it's called. Previously this would not work on RPC calls, but now just works out of the box.

export class MyAgent extends Agent {
	async suggestReply(message) {
		// getCurrentAgent() now correctly works, even when called inside an RPC method
		const { agent } = getCurrentAgent()!;
		return generateText({
			prompt: `Suggest a reply to: "${message}" from "${agent.name}"`,
			tools: [replyWithEmoji],
		});
	}
}

Try it out and tell us what you build!

Agents, Workers - Cloudflare Sandbox SDK adds streaming, code interpreter, Git support, process control and more

Tue, 05 Aug 2025 00:00:00 GMT

We’ve shipped a major release for the @cloudflare/sandbox ↗ SDK, turning it into a full-featured, container-based execution platform that runs securely on Cloudflare Workers.

This update adds live streaming of output, persistent Python and JavaScript code interpreters with rich output support (charts, tables, HTML, JSON), file system access, Git operations, full background process control, and the ability to expose running services via public URLs.

This makes it ideal for building AI agents, CI runners, cloud REPLs, data analysis pipelines, or full developer tools — all without managing infrastructure.

Code interpreter (Python, JS, TS)

Create persistent code contexts with support for rich visual + structured outputs.

createCodeContext(options)

Creates a new code execution context with persistent state.

// Create a Python context
const pythonCtx = await sandbox.createCodeContext({ language: "python" });

// Create a JavaScript context
const jsCtx = await sandbox.createCodeContext({ language: "javascript" });

Options:

language: Programming language ('python' | 'javascript' | 'typescript')
cwd: Working directory (default: /workspace)
envVars: Environment variables for the context

runCode(code, options)

Executes code with optional streaming callbacks.

// Simple execution
const execution = await sandbox.runCode('print("Hello World")', {
	context: pythonCtx,
});

// With streaming callbacks
await sandbox.runCode(
	`
for i in range(5):
    print(f"Step {i}")
    time.sleep(1)
`,
	{
		context: pythonCtx,
		onStdout: (output) => console.log("Real-time:", output.text),
		onResult: (result) => console.log("Result:", result),
	},
);

Options:

language: Programming language ('python' | 'javascript' | 'typescript')
cwd: Working directory (default: /workspace)
envVars: Environment variables for the context

Real-time streaming output

Returns a streaming response for real-time processing.

const stream = await sandbox.runCodeStream(
	"import time; [print(i) for i in range(10)]",
);
// Process the stream as needed

Rich output handling

Interpreter outputs are auto-formatted and returned in multiple formats:

text
html (e.g., Pandas tables)
png, svg (e.g., Matplotlib charts)
json (structured data)
chart (parsed visualizations)

const result = await sandbox.runCode(
	`
import seaborn as sns
import matplotlib.pyplot as plt

data = sns.load_dataset("flights")
pivot = data.pivot("month", "year", "passengers")
sns.heatmap(pivot, annot=True, fmt="d")
plt.title("Flight Passengers")
plt.show()

pivot.to_dict()
`,
	{ context: pythonCtx },
);

if (result.png) {
	console.log("Chart output:", result.png);
}

Preview URLs from Exposed Ports

Start background processes and expose them with live URLs.

await sandbox.startProcess("python -m http.server 8000");
const preview = await sandbox.exposePort(8000);

console.log("Live preview at:", preview.url);

Full process lifecycle control

Start, inspect, and terminate long-running background processes.

const process = await sandbox.startProcess("node server.js");
console.log(`Started process ${process.id} with PID ${process.pid}`);

// Monitor the process
const logStream = await sandbox.streamProcessLogs(process.id);
for await (const log of parseSSEStream<LogEvent>(logStream)) {
	console.log(`Server: ${log.data}`);
}

listProcesses() - List all running processes
getProcess(id) - Get detailed process status
killProcess(id, signal) - Terminate specific processes
killAllProcesses() - Kill all processes
streamProcessLogs(id, options) - Stream logs from running processes
getProcessLogs(id) - Get accumulated process output

Git integration

Clone Git repositories directly into the sandbox.

await sandbox.gitCheckout("https://github.com/user/repo", {
	branch: "main",
	targetDir: "my-project",
});

Sandboxes are still experimental. We're using them to explore how isolated, container-like workloads might scale on Cloudflare — and to help define the developer experience around them.

Agents, Workers AI - OpenAI open models now available on Workers AI

Tue, 05 Aug 2025 00:00:00 GMT

We're thrilled to be a Day 0 partner with OpenAI ↗ to bring their latest open models ↗ to Workers AI, including support for Responses API, Code Interpreter, and Web Search (coming soon).

Get started with the new models at @cf/openai/gpt-oss-120b and @cf/openai/gpt-oss-20b. Check out the blog ↗ for more details about the new models, and the gpt-oss-120b and gpt-oss-20b model pages for more information about pricing and context windows.

Responses API

If you call the model through:

Workers Binding, it will accept/return Responses API – env.AI.run(“@cf/openai/gpt-oss-120b”)
REST API on /run endpoint, it will accept/return Responses API – https://api.cloudflare.com/client/v4/accounts/<account_id>/ai/run/@cf/openai/gpt-oss-120b
REST API on new /responses endpoint, it will accept/return Responses API – https://api.cloudflare.com/client/v4/accounts/<account_id>/ai/v1/responses
REST API for OpenAI Compatible endpoint, it will return Chat Completions (coming soon) – https://api.cloudflare.com/client/v4/accounts/<account_id>/ai/v1/chat/completions

curl https://api.cloudflare.com/client/v4/accounts/<account_id>/ai/v1/responses \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $CLOUDFLARE_API_KEY" \
  -d '{
    "model": "@cf/openai/gpt-oss-120b",
    "reasoning": {"effort": "medium"},
    "input": [
      {
        "role": "user",
        "content": "What are the benefits of open-source models?"
      }
    ]
  }'

Code Interpreter

The model is natively trained to support stateful code execution, and we've implemented support for this feature using our Sandbox SDK ↗ and Containers ↗. Cloudflare's Developer Platform is uniquely positioned to support this feature, so we're very excited to bring our products together to support this new use case.

Web Search (coming soon)

We are working to implement Web Search for the model, where users can bring their own Exa API Key so the model can browse the Internet.

Agents, Workers, Workflows - Run AI-generated code on-demand with Code Sandboxes (new)

Wed, 25 Jun 2025 00:00:00 GMT

AI is supercharging app development for everyone, but we need a safe way to run untrusted, LLM-written code. We’re introducing Sandboxes ↗, which let your Worker run actual processes in a secure, container-based environment.

import { getSandbox } from "@cloudflare/sandbox";
export { Sandbox } from "@cloudflare/sandbox";

export default {
	async fetch(request: Request, env: Env) {
		const sandbox = getSandbox(env.Sandbox, "my-sandbox");
		return sandbox.exec("ls", ["-la"]);
	},
};

Methods

exec(command: string, args: string[], options?: { stream?: boolean }):Execute a command in the sandbox.
gitCheckout(repoUrl: string, options: { branch?: string; targetDir?: string; stream?: boolean }): Checkout a git repository in the sandbox.
mkdir(path: string, options: { recursive?: boolean; stream?: boolean }): Create a directory in the sandbox.
writeFile(path: string, content: string, options: { encoding?: string; stream?: boolean }): Write content to a file in the sandbox.
readFile(path: string, options: { encoding?: string; stream?: boolean }): Read content from a file in the sandbox.
deleteFile(path: string, options?: { stream?: boolean }): Delete a file from the sandbox.
renameFile(oldPath: string, newPath: string, options?: { stream?: boolean }): Rename a file in the sandbox.
moveFile(sourcePath: string, destinationPath: string, options?: { stream?: boolean }): Move a file from one location to another in the sandbox.
ping(): Ping the sandbox.

Sandboxes are still experimental. We're using them to explore how isolated, container-like workloads might scale on Cloudflare — and to help define the developer experience around them.

You can try it today from your Worker, with just a few lines of code. Let us know what you build.

Agents, Workers - Build MCP servers with the Agents SDK

Mon, 07 Apr 2025 00:00:00 GMT

The Agents SDK now includes built-in support for building remote MCP (Model Context Protocol) servers directly as part of your Agent. This allows you to easily create and manage MCP servers, without the need for additional infrastructure or configuration.

The SDK includes a new MCPAgent class that extends the Agent class and allows you to expose resources and tools over the MCP protocol, as well as authorization and authentication to enable remote MCP servers.

export class MyMCP extends McpAgent {
	server = new McpServer({
		name: "Demo",
		version: "1.0.0",
	});

	async init() {
		this.server.resource(`counter`, `mcp://resource/counter`, (uri) => {
			// ...
		});

		this.server.tool(
			"add",
			"Add two numbers together",
			{ a: z.number(), b: z.number() },
			async ({ a, b }) => {
				// ...
			},
		);
	}
}

export class MyMCP extends McpAgent<Env> {
	server = new McpServer({
		name: "Demo",
		version: "1.0.0",
	});

	async init() {
		this.server.resource(`counter`, `mcp://resource/counter`, (uri) => {
			// ...
		});

		this.server.tool(
			"add",
			"Add two numbers together",
			{ a: z.number(), b: z.number() },
			async ({ a, b }) => {
				// ...
			},
		);
	}
}

See the example ↗ for the full code and as the basis for building your own MCP servers, and the client example ↗ for how to build an Agent that acts as an MCP client.

To learn more, review the announcement blog ↗ as part of Developer Week 2025.

Agents SDK updates

We've made a number of improvements to the Agents SDK, including:

Support for building MCP servers with the new MCPAgent class.
The ability to export the current agent, request and WebSocket connection context using import { context } from "agents", allowing you to minimize or avoid direct dependency injection when calling tools.
Fixed a bug that prevented query parameters from being sent to the Agent server from the useAgent React hook.
Automatically converting the agent name in useAgent or useAgentChat to kebab-case to ensure it matches the naming convention expected by routeAgentRequest.

To install or update the Agents SDK, run npm i agents@latest in an existing project, or explore the agents-starter project:

npm create cloudflare@latest -- --template cloudflare/agents-starter

See the full release notes and changelog on the Agents SDK repository ↗ and

Agents, Workers - npm i agents

Tue, 18 Mar 2025 00:00:00 GMT

`agents-sdk` -> `agents` Updated

📝 We've renamed the Agents package to agents!

If you've already been building with the Agents SDK, you can update your dependencies to use the new package name, and replace references to agents-sdk with agents:

# Install the new package
npm i agents

# Remove the old (deprecated) package
npm uninstall agents-sdk

# Find instances of the old package name in your codebase
grep -r 'agents-sdk' .
# Replace instances of the old package name with the new one
# (or use find-replace in your editor)
sed -i 's/agents-sdk/agents/g' $(grep -rl 'agents-sdk' .)

All future updates will be pushed to the new agents package, and the older package has been marked as deprecated.

Agents SDK updates New

We've added a number of big new features to the Agents SDK over the past few weeks, including:

You can now set cors: true when using routeAgentRequest to return permissive default CORS headers to Agent responses.
The regular client now syncs state on the agent (just like the React version).
useAgentChat bug fixes for passing headers/credentials, including properly clearing cache on unmount.
Experimental /schedule module with a prompt/schema for adding scheduling to your app (with evals!).
Changed the internal zod schema to be compatible with the limitations of Google's Gemini models by removing the discriminated union, allowing you to use Gemini models with the scheduling API.

We've also fixed a number of bugs with state synchronization and the React hooks.

// via https://github.com/cloudflare/agents/tree/main/examples/cross-domain
export default {
	async fetch(request, env) {
		return (
			// Set { cors: true } to enable CORS headers.
			(await routeAgentRequest(request, env, { cors: true })) ||
			new Response("Not found", { status: 404 })
		);
	},
};

// via https://github.com/cloudflare/agents/tree/main/examples/cross-domain
export default {
	async fetch(request: Request, env: Env) {
		return (
			// Set { cors: true } to enable CORS headers.
			(await routeAgentRequest(request, env, { cors: true })) ||
			new Response("Not found", { status: 404 })
		);
	},
} satisfies ExportedHandler<Env>;

Call Agent methods from your client code New

We've added a new @unstable_callable() decorator for defining methods that can be called directly from clients. This allows you call methods from within your client code: you can call methods (with arguments) and get native JavaScript objects back.

// server.ts
import { unstable_callable, Agent } from "agents";

export class Rpc extends Agent {
	// Use the decorator to define a callable method
	@unstable_callable({
		description: "rpc test",
	})
	async getHistory() {
		return this.sql`SELECT * FROM history ORDER BY created_at DESC LIMIT 10`;
	}
}

// server.ts
import { unstable_callable, Agent, type StreamingResponse } from "agents";
import type { Env } from "../server";

export class Rpc extends Agent<Env> {
	// Use the decorator to define a callable method
	@unstable_callable({
		description: "rpc test",
	})
	async getHistory() {
		return this.sql`SELECT * FROM history ORDER BY created_at DESC LIMIT 10`;
	}
}

// client.tsx
const { call } = useAgent({ agent: "rpc" });

const fetchUserHistory = async () => {
	try {
		setLoading(true);
		// Call methods directly on the Agent!
		const result = await call("getHistory");
		addToast(`RPC result: ${result}`, "success");
	} catch (error) {
		addToast(`Error: ${error}`, "error");
	} finally {
		setLoading(false);
	}
};

agents-starter Updated

We've fixed a number of small bugs in the agents-starter ↗ project — a real-time, chat-based example application with tool-calling & human-in-the-loop built using the Agents SDK. The starter has also been upgraded to use the latest wrangler v4 release.

If you're new to Agents, you can install and run the agents-starter project in two commands:

# Install it
$ npm create cloudflare@latest agents-starter -- --template="cloudflare/agents-starter"
# Run it
$ npm run start

You can use the starter as a template for your own Agents projects: open up src/server.ts and src/client.tsx to see how the Agents SDK is used.

Agents, Workers - Introducing the Agents SDK

Tue, 25 Feb 2025 00:00:00 GMT

We've released the Agents SDK ↗, a package and set of tools that help you build and ship AI Agents.

You can get up and running with a chat-based AI Agent ↗ (and deploy it to Workers) that uses the Agents SDK, tool calling, and state syncing with a React-based front-end by running the following command:

npm create cloudflare@latest agents-starter -- --template="cloudflare/agents-starter"
# open up README.md and follow the instructions

You can also add an Agent to any existing Workers application by installing the agents package directly

npm i agents

... and then define your first Agent:

import { Agent } from "agents";

export class YourAgent extends Agent<Env> {
	// Build it out
	// Access state on this.state or query the Agent's database via this.sql
	// Handle WebSocket events with onConnect and onMessage
	// Run tasks on a schedule with this.schedule
	// Call AI models
	// ... and/or call other Agents.
}

Head over to the Agents documentation to learn more about the Agents SDK, the SDK APIs, as well as how to test and deploying agents to production.

Agents, Workers, Workflows - Build AI Agents with Example Prompts

Fri, 14 Feb 2025 00:00:00 GMT

We've added an example prompt to help you get started with building AI agents and applications on Cloudflare Workers, including Workflows, Durable Objects, and Workers KV.

You can use this prompt with your favorite AI model, including Claude 3.5 Sonnet, OpenAI's o3-mini, Gemini 2.0 Flash, or Llama 3.3 on Workers AI. Models with large context windows will allow you to paste the prompt directly: provide your own prompt within the <user_prompt></user_prompt> tags.

{paste_prompt_here}
<user_prompt>
user: Build an AI agent using Cloudflare Workflows. The Workflow should run when a new GitHub issue is opened on a specific project with the label 'help' or 'bug', and attempt to help the user troubleshoot the issue by calling the OpenAI API with the issue title and description, and a clear, structured prompt that asks the model to suggest 1-3 possible solutions to the issue. Any code snippets should be formatted in Markdown code blocks. Documentation and sources should be referenced at the bottom of the response. The agent should then post the response to the GitHub issue. The agent should run as the provided GitHub bot account.
</user_prompt>

This prompt is still experimental, but we encourage you to try it out and provide feedback ↗.

Cloudflare changelogs | Agents

Agents, Workers - Agent traces for Think, Flue, and AI SDK instrumented by Agents SDK

Agents, Workers - Preview: @cloudflare/computer agent runtime

Agents, Workers - Cloudflare MCP servers support the new MCP 2026-07-28 Specification

Agents, Workers - Agents SDK adds MCP Specification 2026-07-28 support

Client support

Run stateless servers

Backward compatibility

Migrate existing SDK v1 servers

Deprecations in v0.20.0

Agents - Agents SDK packages support AI SDK v6 and v7

Agents, Workers - Agents SDK reduces MCP schema conversion, adds exposure controls for MCP in Think and Code Mode SDK adds direct host APIs

Control direct MCP tool exposure in Think

Invoke the Code Mode runtime without the AI SDK

Upgrade

Agents, Workers - Agents can respond to MCP elicitation requests

Upgrade

Agents, Workers - Agents SDK adds background sub-agents and a unified turn entry point

Background sub-agents with progress and milestones

One entry point for turns: runTurn

Recovery and reliability

Other improvements

Upgrade

Agents, Workers - Agents SDK improves browser automation, code execution, and recovery

Safer browser automation

Resumable code execution with approvals

Better Think delegation

Voice output device selection

Reliability fixes

Upgrade

Workers, Agents, Workers AI - Introducing GLM-5.2 on Workers AI

Agents, Workers - Agents SDK v0.14.0: Agent Skills, messengers, scheduled tasks, Workflows, and hardened chat recovery

Agent Skills (experimental)

Messengers

Scheduled tasks

Think Workflows

Production hardening for durable chat recovery

MCP transport improvements

Other improvements

Upgrade

Agents - Share sandbox previews through Cloudflare Tunnel

Named tunnels

Upgrade

Agents, Workers - Agents SDK v0.12.4: chat recovery, routing retries, durable Think submissions, and Voice connection control

Chat recovery improvements

Agent state and routing fixes

Durable Think submissions

Think message pruning behavior change

Voice agent connection control

Other improvements

Upgrade

Agents - Agent Lee adds Write Operations and Generative UI

Agent Lee adds Write Operations and Generative UI

Take action with Write Operations

Visualize data with Generative UI

Availability

Containers, Agents - Secure credential injection and dynamic egress policies for Sandboxes

Credential injection

TLS interception

Allow and deny hosts

Dynamic outbound handlers

Get started

Agents, Workers - Agents SDK v0.8.0: readable state, idempotent schedules, typed AgentClient, and Zod 4

Readable state on useAgent and AgentClient

Idempotent schedule()

Typed AgentClient with call inference and stub proxy

@cloudflare/ai-chat fixes

keepAlive() and keepAliveWhile() are no longer experimental

@cloudflare/codemode: TanStack AI integration

Upgrade

Agents, Workers - @cloudflare/codemode v0.2.1: MCP barrel export, zero-dependency main entry point, and custom sandbox modules

New @cloudflare/codemode/mcp export

Zero-dependency main entry point

Custom sandbox modules

Internal normalization and sanitization

Upgrade

Agents - Real-time file watching in Sandboxes

sandbox.watch(path, options)

Server-side consumption with parseSSEStream

Options

One entry point for turns: `runTurn`

Readable `state` on `useAgent` and `AgentClient`

Idempotent `schedule()`

Typed `AgentClient` with `call` inference and `stub` proxy

`@cloudflare/ai-chat` fixes

`keepAlive()` and `keepAliveWhile()` are no longer experimental

`@cloudflare/codemode`: TanStack AI integration

New `@cloudflare/codemode/mcp` export

`sandbox.watch(path, options)`

Server-side consumption with `parseSSEStream`

`keepAlive()` and `keepAliveWhile()`

`waitForMcpConnections`

`@cloudflare/ai-chat` fixes

`DynamicWorkerExecutor`

`@cloudflare/ai-chat` v0.1.0

Deprecate `onStateUpdate` in favor of `onStateChanged`

`sandbox.terminal(request)`