branch:
CHANGELOG.md
39153 bytesRaw
# @cloudflare/ai-chat
## 0.1.9
### Patch Changes
- [#1116](https://github.com/cloudflare/agents/pull/1116) [`dcc3bb2`](https://github.com/cloudflare/agents/commit/dcc3bb20bc1a3ea27dc7d0e333c8b1dc657999b7) Thanks [@whoiskatrin](https://github.com/whoiskatrin)! - Preserve the existing assistant message during tool approval continuations so live client state does not render duplicate assistant messages or tool parts
## 0.1.8
### Patch Changes
- [#1059](https://github.com/cloudflare/agents/pull/1059) [`d0812f7`](https://github.com/cloudflare/agents/commit/d0812f71180d29c48aa4ccc854e14d2ed8517289) Thanks [@threepointone](https://github.com/threepointone)! - Server now responds with `CF_AGENT_STREAM_RESUME_NONE` when a client sends `CF_AGENT_STREAM_RESUME_REQUEST` and no active stream exists. This collapses the previous 5-second timeout to a single WebSocket round-trip, fixing the UI stall on every conversation open/switch/refresh when there is no active stream.
## 0.1.7
### Patch Changes
- [#1046](https://github.com/cloudflare/agents/pull/1046) [`2cde136`](https://github.com/cloudflare/agents/commit/2cde13660a1231a9a14bc50cacf8485af9a07378) Thanks [@threepointone](https://github.com/threepointone)! - Add `agent` and `name` fields to observability events, identifying which agent class and instance emitted each event.
New events: `disconnect` (WebSocket close), `email:receive`, `email:reply`, `queue:create`, and a new `agents:email` channel.
Make `_emit` protected so subclasses can use it. Update `AIChatAgent` to use `_emit` so message/tool events carry agent identity.
## 0.1.6
### Patch Changes
- [#1040](https://github.com/cloudflare/agents/pull/1040) [`766f20b`](https://github.com/cloudflare/agents/commit/766f20bd0b1d7add65fe3522b06b7124d4f8df6c) Thanks [@threepointone](https://github.com/threepointone)! - Changed `waitForMcpConnections` default from `false` to `{ timeout: 10_000 }`. MCP connections are now waited on by default with a 10-second timeout, so `getAITools()` returns the full set of tools in `onChatMessage` without requiring explicit opt-in. Set `waitForMcpConnections = false` to restore the previous behavior.
- [#1020](https://github.com/cloudflare/agents/pull/1020) [`70ebb05`](https://github.com/cloudflare/agents/commit/70ebb05823b48282e3d9e741ab74251c1431ebdd) Thanks [@threepointone](https://github.com/threepointone)! - udpate dependencies
- [#1013](https://github.com/cloudflare/agents/pull/1013) [`11aaaff`](https://github.com/cloudflare/agents/commit/11aaaffb89c375eba9bedf97074ced556dcdd0e7) Thanks [@threepointone](https://github.com/threepointone)! - Fix Gemini "missing thought_signature" error when using client-side tools with `addToolOutput`.
The server-side message builder (`applyChunkToParts`) was dropping `providerMetadata` from tool-input stream chunks instead of storing it as `callProviderMetadata` on tool UIMessage parts. When `convertToModelMessages` later read the persisted messages for the continuation call, `callProviderMetadata` was undefined, so Gemini never received its `thought_signature` back and rejected the request.
- Preserve `callProviderMetadata` (mapped from stream `providerMetadata`) on tool parts in `tool-input-start`, `tool-input-available`, and `tool-input-error` handlers — both create and update paths
- Preserve `providerExecuted` on tool parts (used by `convertToModelMessages` for provider-executed tools like Gemini code execution)
- Preserve `title` on tool parts (tool display name)
- Add `providerExecuted` to `StreamChunkData` type explicitly
- Add 13 regression tests covering all affected codepaths
- [#989](https://github.com/cloudflare/agents/pull/989) [`8404954`](https://github.com/cloudflare/agents/commit/8404954029a62244a87ec38691639f5b8ce9e615) Thanks [@threepointone](https://github.com/threepointone)! - Fix active streams losing UI state after reconnect and dead streams after DO hibernation.
- Send `replayComplete` signal after replaying stored chunks for live streams, so the client flushes accumulated parts to React state immediately instead of waiting for the next live chunk.
- Detect orphaned streams (restored from SQLite after hibernation with no live LLM reader) via `_isLive` flag on `ResumableStream`. On reconnect, send `done: true`, complete the stream, and reconstruct/persist the partial assistant message from stored chunks.
- Client-side: flush `activeStreamRef` on `replayComplete` (keeps stream alive for subsequent live chunks) and on `done` during replay (finalizes orphaned streams).
- [#996](https://github.com/cloudflare/agents/pull/996) [`baf6751`](https://github.com/cloudflare/agents/commit/baf675188c11dded29720842a988a58f8eae2f1b) Thanks [@threepointone](https://github.com/threepointone)! - Fix race condition where MCP tools are intermittently unavailable in onChatMessage after hibernation.
**`agents`**: Added `MCPClientManager.waitForConnections(options?)` which awaits all in-flight connection and discovery operations. Accepts an optional `{ timeout }` in milliseconds. Background restore promises from `restoreConnectionsFromStorage()` are now tracked so callers can wait for them to settle.
**`@cloudflare/ai-chat`**: Added `waitForMcpConnections` opt-in config on `AIChatAgent`. Set to `true` to wait indefinitely, or `{ timeout: 10_000 }` to cap the wait. Default is `false` (non-blocking, preserving existing behavior). For lower-level control, call `this.mcp.waitForConnections()` directly in your `onChatMessage`.
- [#993](https://github.com/cloudflare/agents/pull/993) [`f706e3f`](https://github.com/cloudflare/agents/commit/f706e3f1833d507b53c1ba776982af479ea7cc1b) Thanks [@ferdousbhai](https://github.com/ferdousbhai)! - fix(ai-chat): preserve server tool outputs when client sends approval-responded state
`_mergeIncomingWithServerState` now treats `approval-responded` the same as
`input-available` when the server already has `output-available` for a tool call,
preventing stale client state from overwriting completed tool results.
- [#1038](https://github.com/cloudflare/agents/pull/1038) [`e61cb4a`](https://github.com/cloudflare/agents/commit/e61cb4a5229b4d8ca19202d5633278a87b951df2) Thanks [@threepointone](https://github.com/threepointone)! - fix(ai-chat): preserve server-generated assistant messages when client appends new messages
The `_deleteStaleRows` reconciliation in `persistMessages` now only deletes DB rows when the incoming message set is a subset of the server state (e.g. regenerate trims the conversation). When the client sends new message IDs not yet known to the server, stale deletion is skipped to avoid destroying assistant messages the client hasn't seen.
- [#1014](https://github.com/cloudflare/agents/pull/1014) [`74a3815`](https://github.com/cloudflare/agents/commit/74a3815218f9543a0610d6ccf948fc521b1f788e) Thanks [@threepointone](https://github.com/threepointone)! - Fix `regenerate()` leaving stale assistant messages in SQLite
**Bug 1 — Transport drops `trigger` field:**
`WebSocketChatTransport.sendMessages` was not including the `trigger` field
(e.g. `"regenerate-message"`, `"submit-message"`) in the body payload sent
to the server. The AI SDK passes this field so the server can distinguish
between a new message and a regeneration request. Fixed by adding
`trigger: options.trigger` to the serialized body.
On the server side, `trigger` is now destructured out of the parsed body
alongside `messages` and `clientTools`, so it does not leak into
`options.body` in `onChatMessage`. Users who inspect `options.body` will
not see any change in behavior.
**Bug 2 — `persistMessages` never deletes stale rows:**
`persistMessages` only performed `INSERT ... ON CONFLICT DO UPDATE` (upsert),
so when `regenerate()` removed the last assistant message from the client's
array, the old row persisted in SQLite. On the next `_loadMessagesFromDb`,
the stale assistant message reappeared in `this.messages`, causing:
- Anthropic models to reject with HTTP 400 (conversation must end with a
user message)
- Duplicate/phantom assistant messages across reconnects
Fixed by adding an internal `_deleteStaleRows` option to `persistMessages`.
When the chat-request handler (`CF_AGENT_USE_CHAT_REQUEST`) calls
`persistMessages`, it passes `{ _deleteStaleRows: true }`, which deletes
any DB rows whose IDs are absent from the incoming (post-merge) message set.
This uses the post-merge IDs from `_mergeIncomingWithServerState` to
correctly handle cases where client assistant IDs are remapped to server IDs.
The `_deleteStaleRows` flag is internal only (`@internal` JSDoc) and is
never passed by user code or other handlers (`CF_AGENT_CHAT_MESSAGES`,
`_reply`, `saveMessages`). The default behavior of `persistMessages`
(upsert-only, no deletes) is unchanged.
**Bug 3 — Content-based reconciliation mismatches identical messages:**
`_reconcileAssistantIdsWithServerState` used a single-pass cursor for both
exact-ID and content-based matching. When an exact-ID match jumped the
cursor forward, it skipped server messages needed for content matching
of later identical-text assistant messages (e.g. "Sure", "I understand").
Rewritten with a two-pass approach: Pass 1 resolves all exact-ID matches
and claims server indices. Pass 2 does content-based matching only over
unclaimed server indices. This prevents exact-ID matches from interfering
with content matching, fixing duplicate rows in long conversations with
repeated short assistant responses.
- [#999](https://github.com/cloudflare/agents/pull/999) [`95753da`](https://github.com/cloudflare/agents/commit/95753da49cb68e9e9e486e047b588004163a27fb) Thanks [@threepointone](https://github.com/threepointone)! - Fix `useChat` `status` staying `"ready"` during stream resumption after page refresh.
Four issues prevented stream resumption from working:
1. **addEventListener race:** `onAgentMessage` always handled `CF_AGENT_STREAM_RESUMING` before the transport's listener, bypassing the AI SDK pipeline.
2. **Transport instance instability:** `useMemo` created new transport instances across renders and Strict Mode cycles. When `_pk` changed (async queries, socket recreation), the resolver was stranded on the old transport while `onAgentMessage` called `handleStreamResuming` on the new one.
3. **Chat recreation on `_pk` change:** Using `agent._pk` as the `useChat` `id` caused the AI SDK to recreate the Chat when the socket changed, abandoning the in-flight `makeRequest` (including resume). The resume effect wouldn't re-fire on the new Chat.
4. **Double STREAM_RESUMING:** The server sends `STREAM_RESUMING` from both `onConnect` and the `RESUME_REQUEST` handler, causing duplicate ACKs and double replay without deduplication.
Fixes:
- Replace `addEventListener`-based detection with `handleStreamResuming()` — a synchronous method `onAgentMessage` calls directly, eliminating the race.
- Make the transport a true singleton (`useRef`, created once). Update `transport.agent` every render so sends/listeners always use the latest socket. The resolver survives `_pk` changes because the transport instance never changes.
- Use a stable Chat ID (`initialMessagesCacheKey` based on URL + agent + name) instead of `agent._pk`, preventing Chat recreation on socket changes.
- Add `localRequestIdsRef` guard to skip duplicate `STREAM_RESUMING` messages for streams already handled by the transport.
- [#1029](https://github.com/cloudflare/agents/pull/1029) [`c898308`](https://github.com/cloudflare/agents/commit/c898308d670851e2d79adcc2502f1663ba478b72) Thanks [@threepointone](https://github.com/threepointone)! - Add experimental `keepAlive()` and `keepAliveWhile()` methods to the Agent class. Keeps the Durable Object alive via alarm heartbeats (every 30 seconds), preventing idle eviction during long-running work. `keepAlive()` returns a disposer function; `keepAliveWhile(fn)` runs an async function and automatically cleans up the heartbeat when it completes.
`AIChatAgent` now automatically calls `keepAliveWhile()` during `_reply()` streaming, preventing idle eviction during long LLM generations.
## 0.1.5
### Patch Changes
- [#986](https://github.com/cloudflare/agents/pull/986) [`e0d7a75`](https://github.com/cloudflare/agents/commit/e0d7a75b9c8ef484f7d5fc26f821e575b7a567cb) Thanks [@threepointone](https://github.com/threepointone)! - Terminate the WebSocket chat transport stream when the abort signal fires so
clients exit the "streaming" state after stop/cancel.
## 0.1.4
### Patch Changes
- [#967](https://github.com/cloudflare/agents/pull/967) [`c128447`](https://github.com/cloudflare/agents/commit/c1284478fe212ddd6e1bc915877cee5c10fcfd49) Thanks [@threepointone](https://github.com/threepointone)! - Follow-up to #956. Allow `addToolOutput` to work with tools in `approval-requested` and `approval-responded` states, not just `input-available`. Also adds support for `state: "output-error"` and `errorText` fields, enabling custom denial messages when rejecting tool approvals (addresses remaining items from #955).
Additionally, tool approval rejections (`approved: false`) now auto-continue the conversation when `autoContinue` is set, so the LLM sees the denial and can respond naturally (e.g. suggest alternatives).
This enables the Vercel AI SDK recommended pattern for client-side tool denial:
```ts
addToolOutput({
toolCallId: invocation.toolCallId,
state: "output-error",
errorText: "User declined: insufficient permissions"
});
```
- [#958](https://github.com/cloudflare/agents/pull/958) [`f70a8b9`](https://github.com/cloudflare/agents/commit/f70a8b9e2774d729825b8d85152c403d0c1e6dba) Thanks [@whoiskatrin](https://github.com/whoiskatrin)! - Fix duplicate assistant message persistence when clients resend full history with local assistant IDs that differ from server IDs.
`AIChatAgent.persistMessages()` now reconciles non-tool assistant messages against existing server history by content and order, reusing the server ID instead of inserting duplicate rows.
- [#977](https://github.com/cloudflare/agents/pull/977) [`5426b6f`](https://github.com/cloudflare/agents/commit/5426b6f3a8f394bdbad5e6b5cf22e279249bcdae) Thanks [@dmmulroy](https://github.com/dmmulroy)! - Expose `requestId` in `OnChatMessageOptions` so handlers can send properly-tagged error responses for pre-stream failures.
Also fix `saveMessages()` to pass the full options object (`requestId`, `abortSignal`, `clientTools`, `body`) to `onChatMessage` and use a consistent request ID for `_reply`.
- [#973](https://github.com/cloudflare/agents/pull/973) [`969fbff`](https://github.com/cloudflare/agents/commit/969fbff702d5702c1f0ea6faaecb3dfd0431a01b) Thanks [@threepointone](https://github.com/threepointone)! - Update dependencies
- [#983](https://github.com/cloudflare/agents/pull/983) [`2785f10`](https://github.com/cloudflare/agents/commit/2785f104f187834a0d568ad7db668d961b33707f) Thanks [@threepointone](https://github.com/threepointone)! - Fix abort/cancel support for streaming responses. The framework now properly cancels the reader loop when the abort signal fires and sends a done signal to the client. Added a warning log when cancellation arrives but the stream has not closed (indicating the user forgot to pass `abortSignal` to their LLM call). Also fixed vitest project configs to scope test file discovery and prevent e2e/react tests from being picked up by the wrong runner.
- [#979](https://github.com/cloudflare/agents/pull/979) [`23c90ea`](https://github.com/cloudflare/agents/commit/23c90ea4bdd63c03f28c40e1c3594b34ff523bf7) Thanks [@mattzcarey](https://github.com/mattzcarey)! - Fix jsonSchema not initialized error when calling getAITools() in onChatMessage
- [#980](https://github.com/cloudflare/agents/pull/980) [`00c576d`](https://github.com/cloudflare/agents/commit/00c576de0ddcbac1ae4496abb14804cfb34e251e) Thanks [@threepointone](https://github.com/threepointone)! - Fix `_sanitizeMessageForPersistence` stripping Anthropic `redacted_thinking` blocks. The sanitizer now strips OpenAI ephemeral metadata first, then filters out only reasoning parts that are truly empty (no text and no remaining `providerMetadata`). This preserves Anthropic's `redacted_thinking` blocks (stored as empty-text reasoning parts with `providerMetadata.anthropic.redactedData`) while still removing OpenAI placeholders. Fixes #978.
- [#953](https://github.com/cloudflare/agents/pull/953) [`bd22d60`](https://github.com/cloudflare/agents/commit/bd22d6005fab16d0dc358274dcb12d368a31e076) Thanks [@mattzcarey](https://github.com/mattzcarey)! - Moved `/get-messages` endpoint handling from a prototype `override onRequest()` method to a constructor wrapper. This ensures the endpoint always works, even when users override `onRequest` without calling `super.onRequest()`.
- [#956](https://github.com/cloudflare/agents/pull/956) [`ab401a0`](https://github.com/cloudflare/agents/commit/ab401a0e0b6942490e845cc9e34d9f17f65cbde8) Thanks [@whoiskatrin](https://github.com/whoiskatrin)! - Fix denied tool approvals (`CF_AGENT_TOOL_APPROVAL` with `approved: false`) to transition tool parts to `output-denied` instead of `approval-responded`.
This ensures `convertToModelMessages` emits a `tool_result` for denied approvals, which is required by providers like Anthropic.
Also adds regression tests for denied approval behavior, including rejection from `approval-requested` state.
- [#982](https://github.com/cloudflare/agents/pull/982) [`5a851be`](https://github.com/cloudflare/agents/commit/5a851bef389683a13380626c8bed515a6351b172) Thanks [@threepointone](https://github.com/threepointone)! - Undeprecate client tool APIs (`createToolsFromClientSchemas`, `clientTools`, `AITool`, `extractClientToolSchemas`, and the `tools` option on `useAgentChat`) for SDK/platform use cases where tools are defined dynamically at runtime. Fix spurious `detectToolsRequiringConfirmation` deprecation warning when using the `tools` option.
## 0.1.3
### Patch Changes
- [#954](https://github.com/cloudflare/agents/pull/954) [`943c407`](https://github.com/cloudflare/agents/commit/943c4070992bb836625abb5bf4e3271a6f52f7a2) Thanks [@threepointone](https://github.com/threepointone)! - update dependencies
## 0.1.2
### Patch Changes
- [#930](https://github.com/cloudflare/agents/pull/930) [`cd408fe`](https://github.com/cloudflare/agents/commit/cd408fe495176e67066ebbb3962c224ada702124) Thanks [@threepointone](https://github.com/threepointone)! - Fix stale agent reference in useAgentChat transport under React StrictMode
The `agentRef` was updated via `useEffect` (async, after render), but the `WebSocketChatTransport` is created in `useMemo` (sync, during render). When the agent reconnects or switches, `useMemo` would capture the old (closed) agent because the effect hadn't fired yet — causing `sendMessage` to send to a dead WebSocket. Fixed by updating `agentRef.current` synchronously during render, matching the pattern already used by other refs in the same hook.
## 0.1.1
### Patch Changes
- [`eeadbf4`](https://github.com/cloudflare/agents/commit/eeadbf4e780e2477798185cbc7a8abbeff2eadda) Thanks [@threepointone](https://github.com/threepointone)! - Add @ai-sdk/react as peer dependency to ai-chat
Declare @ai-sdk/react ^3.0.0 as a peerDependency in packages/ai-chat/package.json to express runtime compatibility with the React SDK. package-lock.json was updated to reflect the resulting dependency graph changes.
## 0.1.0
The first minor release of `@cloudflare/ai-chat` — a major step up from the `agents/ai-chat-agent` re-export. This release refactors the internals (extracting ResumableStream, adding a WebSocket ChatTransport, simplifying SSE parsing) and ships a wave of bug fixes for streaming, tool continuations, and message persistence. New features include `maxPersistedMessages` for storage caps, `body` for custom request data, row size protection, incremental persistence, and data parts — typed JSON blobs that can be attached to messages alongside text for citations, progress indicators, and usage metadata. Tool approval (`needsApproval`) now persists across page refresh, client tools survive DO hibernation, and `autoContinueAfterToolResult` defaults to `true` so the LLM responds after tool results without explicit opt-in.
### Minor Changes
- [#899](https://github.com/cloudflare/agents/pull/899) [`04c6411`](https://github.com/cloudflare/agents/commit/04c6411c9a73fe48784d7ce86150d62cf54becda) Thanks [@threepointone](https://github.com/threepointone)! - Refactor AIChatAgent: extract ResumableStream class, add WebSocket ChatTransport, simplify SSE parsing.
**Bug fixes:**
- Fix `setMessages` functional updater sending empty array to server
- Fix `_sendPlaintextReply` creating multiple text parts instead of one
- Fix uncaught exception on empty/invalid request body
- Fix `CF_AGENT_MESSAGE_UPDATED` not broadcast for streaming messages
- Fix stream resumption race condition (client-initiated resume request + replay flag)
- Fix `_streamCompletionPromise` not resolved on error (tool continuations could hang)
- Fix `body` lost during tool continuations (now preserved alongside `clientTools`)
- Fix `clearAll()` not clearing in-memory chunk buffer (orphaned chunks could flush after clear)
- Fix errored streams never cleaned up by garbage collector
- Fix `reasoning-delta` silently dropping data when `reasoning-start` was missed (stream resumption)
- Fix row size guard using `string.length` instead of UTF-8 byte count for SQLite limits
- Fix `completed` guard on abort listener to prevent redundant cancel after stream completion
**New features:**
- `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 with hash-based cache to skip redundant SQL writes
- Row size guard — automatic two-pass compaction when messages approach SQLite 2MB limit
- `onFinish` is now optional — framework handles abort controller cleanup and observability
- Stream chunk size guard in ResumableStream (skip oversized chunks for replay)
- Full tool streaming lifecycle in message-builder (tool-input-start/delta/error, tool-output-error)
**Docs:**
- New `docs/chat-agents.md` — comprehensive AIChatAgent and useAgentChat reference
- Rewritten README, migration guides, human-in-the-loop, resumable streaming, client tools docs
- New `examples/ai-chat/` example with modern patterns and Workers AI
**Deprecations (with console.warn):**
- `createToolsFromClientSchemas()`, `extractClientToolSchemas()`, `detectToolsRequiringConfirmation()`
- `tools`, `toolsRequiringConfirmation`, `experimental_automaticToolResolution` options
- `addToolResult()` (use `addToolOutput()`)
- [#919](https://github.com/cloudflare/agents/pull/919) [`6b6497c`](https://github.com/cloudflare/agents/commit/6b6497c65e07175ffd83f8cf3a6b3371e2dc17eb) Thanks [@threepointone](https://github.com/threepointone)! - Change `autoContinueAfterToolResult` default from `false` to `true`.
Client-side tool results and tool approvals now automatically trigger a server continuation by default, matching the behavior of server-executed tools (which auto-continue via `streamText`'s multi-step). This eliminates the most common setup friction with client tools — the LLM now responds after receiving tool results without requiring explicit opt-in.
To restore the previous behavior, set `autoContinueAfterToolResult: false` in `useAgentChat`.
### Patch Changes
- [#900](https://github.com/cloudflare/agents/pull/900) [`16b2dca`](https://github.com/cloudflare/agents/commit/16b2dcaf5adc152e78f01230dd99d4710867d4b6) Thanks [@deathbyknowledge](https://github.com/deathbyknowledge)! - Add support for `data-*` stream parts (developer-defined typed JSON blobs) in the shared message builder and client hook.
**Data part handling:**
`applyChunkToParts` now handles `data-*` prefixed chunk types, covering both server persistence and client reconstruction (stream resume, cross-tab broadcast). Transient parts (`transient: true`) are broadcast to connected clients but excluded from `message.parts` and SQLite persistence. Non-transient parts support reconciliation by type+id — a second chunk with the same type and id updates the existing part's data in-place instead of appending a duplicate.
**`onData` callback forwarding:**
`useAgentChat` now invokes the `onData` callback for `data-*` chunks on the stream resumption and cross-tab broadcast codepaths, which bypass the AI SDK's internal pipeline. For new messages sent via the transport, the AI SDK already invokes `onData` internally. This is the correct way to consume transient data parts on the client since they are not added to `message.parts`.
- [#922](https://github.com/cloudflare/agents/pull/922) [`c8e5244`](https://github.com/cloudflare/agents/commit/c8e524499d902229e8ac83afd6cf2864f888cecc) Thanks [@threepointone](https://github.com/threepointone)! - Fix tool approval UI not surviving page refresh, and fix invalid prompt error after approval
- Handle `tool-approval-request` and `tool-output-denied` stream chunks in the server-side message builder. Previously these were only handled client-side, so the server never transitioned tool parts to `approval-requested` or `output-denied` state.
- Persist the streaming message to SQLite (without broadcasting) when a tool enters `approval-requested` state. The stream is paused waiting for user approval, so this is a natural persistence point. Without this, refreshing the page would reload from SQLite where the tool was still in `input-available` state, showing "Running..." instead of the Approve/Reject UI.
- On stream completion, update the early-persisted message in place rather than appending a duplicate.
- Fix `_applyToolApproval` to merge with existing approval data instead of replacing it. Previously `approval: { approved }` would overwrite the entire object, losing the `id` field that `convertToModelMessages` needs to produce a valid `tool-approval-request` content part. This caused an `InvalidPromptError` on the continuation stream after approval.
- [#897](https://github.com/cloudflare/agents/pull/897) [`994a808`](https://github.com/cloudflare/agents/commit/994a808abb5620b57aba4e0e0125bbcd89c1ae5f) Thanks [@alexanderjacobsen](https://github.com/alexanderjacobsen)! - Fix client tool schemas lost after DO restart by re-sending them with CF_AGENT_TOOL_RESULT
- [#916](https://github.com/cloudflare/agents/pull/916) [`24e16e0`](https://github.com/cloudflare/agents/commit/24e16e025b82dbd7b321339a18c6d440b2879136) Thanks [@threepointone](https://github.com/threepointone)! - Widen peer dependency ranges across packages to prevent cascading major bumps during 0.x minor releases. Mark `@cloudflare/ai-chat` and `@cloudflare/codemode` as optional peer dependencies of `agents` to fix unmet peer dependency warnings during installation.
- [#912](https://github.com/cloudflare/agents/pull/912) [`baa87cc`](https://github.com/cloudflare/agents/commit/baa87cceccd11ce051af5d2918831ec8eddd86fb) Thanks [@threepointone](https://github.com/threepointone)! - Persist request context across Durable Object hibernation.
Persist `_lastBody` and `_lastClientTools` to SQLite so custom body fields and client tool schemas survive Durable Object hibernation during tool continuation flows (issue #887). Add test coverage for body forwarding during tool auto-continuation, and update JSDoc for `OnChatMessageOptions.body` to document tool continuation and hibernation behavior.
- [#913](https://github.com/cloudflare/agents/pull/913) [`bc91c9a`](https://github.com/cloudflare/agents/commit/bc91c9a63aefa2faf37db2ad7b5f3f382a1de101) Thanks [@threepointone](https://github.com/threepointone)! - Sync `_lastClientTools` cache and SQLite when client tools arrive via `CF_AGENT_TOOL_RESULT`, and align the wire type with `ClientToolSchema` (`JSONSchema7` instead of `Record<string, unknown>`)
- [#919](https://github.com/cloudflare/agents/pull/919) [`6b6497c`](https://github.com/cloudflare/agents/commit/6b6497c65e07175ffd83f8cf3a6b3371e2dc17eb) Thanks [@threepointone](https://github.com/threepointone)! - Add auto-continuation support for tool approval (`needsApproval`).
When a tool with `needsApproval: true` is approved via `CF_AGENT_TOOL_APPROVAL`, the server can now automatically continue the conversation (matching the existing `autoContinue` behavior of `CF_AGENT_TOOL_RESULT`). The client hook passes `autoContinue` with approval messages when `autoContinueAfterToolResult` is enabled. Also fixes silent data loss where `tool-output-available` events for tool calls in previous assistant messages were dropped during continuation streams by adding a cross-message fallback search in `_streamSSEReply`.
- [#910](https://github.com/cloudflare/agents/pull/910) [`a668155`](https://github.com/cloudflare/agents/commit/a668155598aa8cd2f53b724391d1c538f3e96a2d) Thanks [@threepointone](https://github.com/threepointone)! - Add structural message validation and fix message metadata on broadcast/resume path.
**Structural message validation:**
Messages loaded from SQLite are now validated for required structure (non-empty `id` string, valid `role`, `parts` is an array). Malformed rows — from corruption, manual tampering, or schema drift — are logged with a warning and silently skipped instead of crashing the agent. This is intentionally lenient: empty `parts` arrays are allowed (streams that errored mid-flight), and no tool/data schema validation is performed at load time (that remains a userland concern via `safeValidateUIMessages` from the AI SDK).
**Message metadata on broadcast/resume path:**
The server already captures `messageMetadata` from `start`, `finish`, and `message-metadata` stream chunks and persists it on `message.metadata`. However, the client-side broadcast path (multi-tab sync) and stream resume path (reconnection) did not propagate metadata — the `activeStreamRef` only tracked `parts`. Now it also tracks `metadata`, and `flushActiveStreamToMessages` includes it in the partial message flushed to React state. This means cross-tab clients and reconnecting clients see metadata (model info, token usage, timestamps) during streaming, not just after the final `CF_AGENT_CHAT_MESSAGES` broadcast.
## 0.0.8
### Patch Changes
- [#882](https://github.com/cloudflare/agents/pull/882) [`584cebe`](https://github.com/cloudflare/agents/commit/584cebe882f437a685b96b26b15200dc50ba70e1) Thanks [@alexanderjacobsen](https://github.com/alexanderjacobsen)! - Fix multi-step client tool calling: pass stored client tool schemas to `onChatMessage` during tool continuations so the LLM can call additional client tools after auto-continuation. Also add a re-trigger mechanism to the client-side tool resolution effect to handle tool calls arriving during active resolution.
- [#891](https://github.com/cloudflare/agents/pull/891) [`0723b99`](https://github.com/cloudflare/agents/commit/0723b9909f037d494e0c7db43e031c952578c82e) Thanks [@ask-bonk](https://github.com/apps/ask-bonk)! - Fix `getCurrentAgent()` returning `undefined` connection when used with `@cloudflare/ai-chat` and Vite SSR
Re-export `agentContext` as `__DO_NOT_USE_WILL_BREAK__agentContext` from the main `agents` entry point and update `@cloudflare/ai-chat` to import it from `agents` instead of the `agents/internal_context` subpath export. This prevents Vite SSR pre-bundling from creating two separate `AsyncLocalStorage` instances, which caused `getCurrentAgent().connection` to be `undefined` inside `onChatMessage` and tool `execute` functions.
The `agents/internal_context` subpath export has been removed from `package.json` and the deprecated `agentContext` alias has been removed from `internal_context.ts`. This was never a public API.
- [#886](https://github.com/cloudflare/agents/pull/886) [`4292f6b`](https://github.com/cloudflare/agents/commit/4292f6ba6d49201c88b09553452c3b243620f35b) Thanks [@whoiskatrin](https://github.com/whoiskatrin)! - Forward custom body fields from client requests to `onChatMessage` options
Custom data sent via `prepareSendMessagesRequest` or the AI SDK's `body` option in `sendMessage` is now available in the `onChatMessage` handler through `options.body`. This allows passing dynamic context (e.g., model selection, temperature, custom metadata) from the client to the server without workarounds.
## 0.0.7
### Patch Changes
- [#859](https://github.com/cloudflare/agents/pull/859) [`3de98a3`](https://github.com/cloudflare/agents/commit/3de98a398d55aeca51c7b845ed4c5d6051887d6d) Thanks [@threepointone](https://github.com/threepointone)! - broaden peer deps
- [#865](https://github.com/cloudflare/agents/pull/865) [`c3211d0`](https://github.com/cloudflare/agents/commit/c3211d0b0cc36aa294c15569ae650d3afeab9926) Thanks [@threepointone](https://github.com/threepointone)! - update dependencies
## 0.0.6
### Patch Changes
- [#829](https://github.com/cloudflare/agents/pull/829) [`83f137f`](https://github.com/cloudflare/agents/commit/83f137f7046aeafc3b480b5aa4518f6290b14406) Thanks [@Muhammad-Bin-Ali](https://github.com/Muhammad-Bin-Ali)! - Fix duplicate assistant messages when using needsApproval tools
When calling `addToolApprovalResponse`, the original assistant message is now updated in place instead of creating a duplicate with a new ID.
- Updated dependencies [[`68916bf`](https://github.com/cloudflare/agents/commit/68916bfa08358d4bb5d61aff37acd8dc4ffc950e), [`3f490d0`](https://github.com/cloudflare/agents/commit/3f490d045844e4884db741afbb66ca1fe65d4093)]:
- agents@0.3.10
## 0.0.5
### Patch Changes
- [#813](https://github.com/cloudflare/agents/pull/813) [`7aebab3`](https://github.com/cloudflare/agents/commit/7aebab369d1bef6c685e05a4a3bd6627edcb87db) Thanks [@threepointone](https://github.com/threepointone)! - update dependencies
- [#797](https://github.com/cloudflare/agents/pull/797) [`77be4f8`](https://github.com/cloudflare/agents/commit/77be4f8149e41730148a360adfff9e66becdd5ed) Thanks [@iTrooz](https://github.com/iTrooz)! - refactor(ai-chat): put SSE reply and plaintext reply logic into 2 separate functions
- [#800](https://github.com/cloudflare/agents/pull/800) [`a54edf5`](https://github.com/cloudflare/agents/commit/a54edf56b462856d1ef4f424c2363ac43a53c46e) Thanks [@threepointone](https://github.com/threepointone)! - Update dependencies
- [#818](https://github.com/cloudflare/agents/pull/818) [`7c74336`](https://github.com/cloudflare/agents/commit/7c743360d7e3639e187725391b9d5c114838bd18) Thanks [@threepointone](https://github.com/threepointone)! - update dependencies
- [#795](https://github.com/cloudflare/agents/pull/795) [`99cbca0`](https://github.com/cloudflare/agents/commit/99cbca0847d0d6c97f44b73f2eb155dabe590032) Thanks [@Jerrynh770](https://github.com/Jerrynh770)! - Fix resumable streaming to avoid delivering live chunks before resume ACK
- Updated dependencies [[`0c3c9bb`](https://github.com/cloudflare/agents/commit/0c3c9bb62ceff66ed38d3bbd90c767600f1f3453), [`0c3c9bb`](https://github.com/cloudflare/agents/commit/0c3c9bb62ceff66ed38d3bbd90c767600f1f3453), [`d1a0c2b`](https://github.com/cloudflare/agents/commit/d1a0c2b73b1119d71e120091753a6bcca0e2faa9), [`6218541`](https://github.com/cloudflare/agents/commit/6218541e9c1e40ccbaa25b2d9d93858c0ad81ffa), [`6218541`](https://github.com/cloudflare/agents/commit/6218541e9c1e40ccbaa25b2d9d93858c0ad81ffa), [`6218541`](https://github.com/cloudflare/agents/commit/6218541e9c1e40ccbaa25b2d9d93858c0ad81ffa), [`6218541`](https://github.com/cloudflare/agents/commit/6218541e9c1e40ccbaa25b2d9d93858c0ad81ffa), [`fd79481`](https://github.com/cloudflare/agents/commit/fd7948180abf066fa3d27911a83ffb4c91b3f099), [`6218541`](https://github.com/cloudflare/agents/commit/6218541e9c1e40ccbaa25b2d9d93858c0ad81ffa), [`0c3c9bb`](https://github.com/cloudflare/agents/commit/0c3c9bb62ceff66ed38d3bbd90c767600f1f3453), [`6218541`](https://github.com/cloudflare/agents/commit/6218541e9c1e40ccbaa25b2d9d93858c0ad81ffa), [`6218541`](https://github.com/cloudflare/agents/commit/6218541e9c1e40ccbaa25b2d9d93858c0ad81ffa), [`e20da53`](https://github.com/cloudflare/agents/commit/e20da5319eb46bac6ac580edf71836b00ac6f8bb), [`f604008`](https://github.com/cloudflare/agents/commit/f604008957f136241815909319a552bad6738b58), [`7aebab3`](https://github.com/cloudflare/agents/commit/7aebab369d1bef6c685e05a4a3bd6627edcb87db), [`a54edf5`](https://github.com/cloudflare/agents/commit/a54edf56b462856d1ef4f424c2363ac43a53c46e), [`7c74336`](https://github.com/cloudflare/agents/commit/7c743360d7e3639e187725391b9d5c114838bd18), [`6218541`](https://github.com/cloudflare/agents/commit/6218541e9c1e40ccbaa25b2d9d93858c0ad81ffa), [`ded8d3e`](https://github.com/cloudflare/agents/commit/ded8d3e8aeba0358ebd4aecb5ba15344b5a21db1)]:
- agents@0.3.7
## 0.0.4
### Patch Changes
- [#761](https://github.com/cloudflare/agents/pull/761) [`0e8fc1e`](https://github.com/cloudflare/agents/commit/0e8fc1e8cca3ad5acb51f5a0c92528c5b6beb358) Thanks [@iTrooz](https://github.com/iTrooz)! - Allow returning a non-streaming reponse from onChatMessage()
- [#771](https://github.com/cloudflare/agents/pull/771) [`87dc96d`](https://github.com/cloudflare/agents/commit/87dc96d19de1d26dbb2badecbb9955a4eb8e9e2e) Thanks [@threepointone](https://github.com/threepointone)! - update dependencies
- Updated dependencies [[`cf8a1e7`](https://github.com/cloudflare/agents/commit/cf8a1e7a24ecaac62c2aefca7b0fd5bf1373e8bd), [`87dc96d`](https://github.com/cloudflare/agents/commit/87dc96d19de1d26dbb2badecbb9955a4eb8e9e2e)]:
- agents@0.3.4
## 0.0.3
### Patch Changes
- [`a5d0137`](https://github.com/cloudflare/agents/commit/a5d01379b9ad2d88bc028c50f1858b4e69f106c5) Thanks [@threepointone](https://github.com/threepointone)! - trigger a new release
- Updated dependencies [[`a5d0137`](https://github.com/cloudflare/agents/commit/a5d01379b9ad2d88bc028c50f1858b4e69f106c5)]:
- agents@0.3.3
## 0.0.2
### Patch Changes
- [#756](https://github.com/cloudflare/agents/pull/756) [`0c4275f`](https://github.com/cloudflare/agents/commit/0c4275f8f4b71c264c32c3742d151ef705739c2f) Thanks [@threepointone](https://github.com/threepointone)! - feat: split ai-chat and codemode into separate packages
Extract @cloudflare/ai-chat and @cloudflare/codemode into their own packages
with comprehensive READMEs. Update agents README to remove chat-specific
content and point to new packages. Fix documentation imports to reflect
new package structure.
Maintains backward compatibility, no breaking changes.
- Updated dependencies [[`0c4275f`](https://github.com/cloudflare/agents/commit/0c4275f8f4b71c264c32c3742d151ef705739c2f), [`f12553f`](https://github.com/cloudflare/agents/commit/f12553f2fa65912c68d9a7620b9a11b70b8790a2)]:
- agents@0.3.2