@mastra/memory 1.0.0-beta.1 → 1.0.0-beta.11

package/CHANGELOG.md

@@ -1,5 +1,360 @@
 # @mastra/memory
 
+## 1.0.0-beta.11
+
+### Patch Changes
+
+- dependencies updates: ([#10133](https://github.com/mastra-ai/mastra/pull/10133))
+  - Updated dependency [`js-tiktoken@^1.0.21` ↗︎](https://www.npmjs.com/package/js-tiktoken/v/1.0.21) (from `^1.0.20`, in `dependencies`)
+
+- Add embedded documentation support for Mastra packages ([#11472](https://github.com/mastra-ai/mastra/pull/11472))
+
+  Mastra packages now include embedded documentation in the published npm package under `dist/docs/`. This enables coding agents and AI assistants to understand and use the framework by reading documentation directly from `node_modules`.
+
+  Each package includes:
+  - **SKILL.md** - Entry point explaining the package's purpose and capabilities
+  - **SOURCE_MAP.json** - Machine-readable index mapping exports to types and implementation files
+  - **Topic folders** - Conceptual documentation organized by feature area
+
+  Documentation is driven by the `packages` frontmatter field in MDX files, which maps docs to their corresponding packages. CI validation ensures all docs include this field.
+
+- Fixed client-side tool invocations not being stored in memory. Previously, tool invocations with state 'call' were filtered out before persistence, which incorrectly removed client-side tools. Now only streaming intermediate states ('partial-call') are filtered. ([#11630](https://github.com/mastra-ai/mastra/pull/11630))
+
+  Fixed a crash when updating working memory with an empty or null update; existing data is now preserved.
+
+- Adds thread cloning to create independent copies of conversations that can diverge. ([#11517](https://github.com/mastra-ai/mastra/pull/11517))
+
+  ```typescript
+  // Clone a thread
+  const { thread, clonedMessages } = await memory.cloneThread({
+    sourceThreadId: 'thread-123',
+    title: 'My Clone',
+    options: {
+      messageLimit: 10, // optional: only copy last N messages
+    },
+  });
+
+  // Check if a thread is a clone
+  if (memory.isClone(thread)) {
+    const source = await memory.getSourceThread(thread.id);
+  }
+
+  // List all clones of a thread
+  const clones = await memory.listClones('thread-123');
+  ```
+
+  Includes:
+  - Storage implementations for InMemory, PostgreSQL, LibSQL, Upstash
+  - API endpoint: `POST /api/memory/threads/:threadId/clone`
+  - Embeddings created for cloned messages (semantic recall)
+  - Clone button in playground UI Memory tab
+
+- Unified `getWorkflowRunById` and `getWorkflowRunExecutionResult` into a single API that returns `WorkflowState` with both metadata and execution state. ([#11429](https://github.com/mastra-ai/mastra/pull/11429))
+
+  **What changed:**
+  - `getWorkflowRunById` now returns a unified `WorkflowState` object containing metadata (runId, workflowName, resourceId, createdAt, updatedAt) along with processed execution state (status, result, error, payload, steps)
+  - Added optional `fields` parameter to request only specific fields for better performance
+  - Added optional `withNestedWorkflows` parameter to control nested workflow step inclusion
+  - Removed `getWorkflowRunExecutionResult` - use `getWorkflowRunById` instead (breaking change)
+  - Removed `/execution-result` API endpoints from server (breaking change)
+  - Removed `runExecutionResult()` method from client SDK (breaking change)
+  - Removed `GetWorkflowRunExecutionResultResponse` type from client SDK (breaking change)
+
+  **Before:**
+
+  ```typescript
+  // Had to call two different methods for different data
+  const run = await workflow.getWorkflowRunById(runId); // Returns raw WorkflowRun with snapshot
+  const result = await workflow.getWorkflowRunExecutionResult(runId); // Returns processed execution state
+  ```
+
+  **After:**
+
+  ```typescript
+  // Single method returns everything
+  const run = await workflow.getWorkflowRunById(runId);
+  // Returns: { runId, workflowName, resourceId, createdAt, updatedAt, status, result, error, payload, steps }
+
+  // Request only specific fields for better performance (avoids expensive step fetching)
+  const status = await workflow.getWorkflowRunById(runId, { fields: ['status'] });
+
+  // Skip nested workflow steps for faster response
+  const run = await workflow.getWorkflowRunById(runId, { withNestedWorkflows: false });
+  ```
+
+  **Why:** The previous API required calling two separate methods to get complete workflow run information. This unification simplifies the API surface and gives users control over performance - fetching all steps (especially nested workflows) can be expensive, so the `fields` and `withNestedWorkflows` options let users request only what they need.
+
+- Updated dependencies [[`d2d3e22`](https://github.com/mastra-ai/mastra/commit/d2d3e22a419ee243f8812a84e3453dd44365ecb0), [`bc72b52`](https://github.com/mastra-ai/mastra/commit/bc72b529ee4478fe89ecd85a8be47ce0127b82a0), [`05b8bee`](https://github.com/mastra-ai/mastra/commit/05b8bee9e50e6c2a4a2bf210eca25ee212ca24fa), [`c042bd0`](https://github.com/mastra-ai/mastra/commit/c042bd0b743e0e86199d0cb83344ca7690e34a9c), [`940a2b2`](https://github.com/mastra-ai/mastra/commit/940a2b27480626ed7e74f55806dcd2181c1dd0c2), [`e0941c3`](https://github.com/mastra-ai/mastra/commit/e0941c3d7fc75695d5d258e7008fd5d6e650800c), [`0c0580a`](https://github.com/mastra-ai/mastra/commit/0c0580a42f697cd2a7d5973f25bfe7da9055038a), [`28f5f89`](https://github.com/mastra-ai/mastra/commit/28f5f89705f2409921e3c45178796c0e0d0bbb64), [`e601b27`](https://github.com/mastra-ai/mastra/commit/e601b272c70f3a5ecca610373aa6223012704892), [`3d3366f`](https://github.com/mastra-ai/mastra/commit/3d3366f31683e7137d126a3a57174a222c5801fb), [`5a4953f`](https://github.com/mastra-ai/mastra/commit/5a4953f7d25bb15ca31ed16038092a39cb3f98b3), [`eb9e522`](https://github.com/mastra-ai/mastra/commit/eb9e522ce3070a405e5b949b7bf5609ca51d7fe2), [`20e6f19`](https://github.com/mastra-ai/mastra/commit/20e6f1971d51d3ff6dd7accad8aaaae826d540ed), [`4f0b3c6`](https://github.com/mastra-ai/mastra/commit/4f0b3c66f196c06448487f680ccbb614d281e2f7), [`74c4f22`](https://github.com/mastra-ai/mastra/commit/74c4f22ed4c71e72598eacc346ba95cdbc00294f), [`81b6a8f`](https://github.com/mastra-ai/mastra/commit/81b6a8ff79f49a7549d15d66624ac1a0b8f5f971), [`e4d366a`](https://github.com/mastra-ai/mastra/commit/e4d366aeb500371dd4210d6aa8361a4c21d87034), [`a4f010b`](https://github.com/mastra-ai/mastra/commit/a4f010b22e4355a5fdee70a1fe0f6e4a692cc29e), [`73b0bb3`](https://github.com/mastra-ai/mastra/commit/73b0bb394dba7c9482eb467a97ab283dbc0ef4db), [`5627a8c`](https://github.com/mastra-ai/mastra/commit/5627a8c6dc11fe3711b3fa7a6ffd6eb34100a306), [`3ff45d1`](https://github.com/mastra-ai/mastra/commit/3ff45d10e0c80c5335a957ab563da72feb623520), [`251df45`](https://github.com/mastra-ai/mastra/commit/251df4531407dfa46d805feb40ff3fb49769f455), [`f894d14`](https://github.com/mastra-ai/mastra/commit/f894d148946629af7b1f452d65a9cf864cec3765), [`c2b9547`](https://github.com/mastra-ai/mastra/commit/c2b9547bf435f56339f23625a743b2147ab1c7a6), [`580b592`](https://github.com/mastra-ai/mastra/commit/580b5927afc82fe460dfdf9a38a902511b6b7e7f), [`58e3931`](https://github.com/mastra-ai/mastra/commit/58e3931af9baa5921688566210f00fb0c10479fa), [`08bb631`](https://github.com/mastra-ai/mastra/commit/08bb631ae2b14684b2678e3549d0b399a6f0561e), [`4fba91b`](https://github.com/mastra-ai/mastra/commit/4fba91bec7c95911dc28e369437596b152b04cd0), [`12b0cc4`](https://github.com/mastra-ai/mastra/commit/12b0cc4077d886b1a552637dedb70a7ade93528c)]:
+  - @mastra/core@1.0.0-beta.20
+
+## 1.0.0-beta.10
+
+### Patch Changes
+
+- Updated dependencies [[`e54953e`](https://github.com/mastra-ai/mastra/commit/e54953ed8ce1b28c0d62a19950163039af7834b4), [`7d56d92`](https://github.com/mastra-ai/mastra/commit/7d56d9213886e8353956d7d40df10045fd12b299), [`fdac646`](https://github.com/mastra-ai/mastra/commit/fdac646033a0930a1a4e00d13aa64c40bb7f1e02), [`d07b568`](https://github.com/mastra-ai/mastra/commit/d07b5687819ea8cb1dffa776d0c1765faf4aa1ae), [`70b300e`](https://github.com/mastra-ai/mastra/commit/70b300ebc631dfc0aa14e61547fef7994adb4ea6), [`68ec97d`](https://github.com/mastra-ai/mastra/commit/68ec97d4c07c6393fcf95c2481fc5d73da99f8c8), [`4aa55b3`](https://github.com/mastra-ai/mastra/commit/4aa55b383cf06043943359ea316572fd969861a7)]:
+  - @mastra/core@1.0.0-beta.19
+  - @mastra/schema-compat@1.0.0-beta.5
+
+## 1.0.0-beta.9
+
+### Patch Changes
+
+- Updated dependencies [[`b5dc973`](https://github.com/mastra-ai/mastra/commit/b5dc9733a5158850298dfb103acb3babdba8a318), [`af56599`](https://github.com/mastra-ai/mastra/commit/af56599d73244ae3bf0d7bcade656410f8ded37b)]:
+  - @mastra/core@1.0.0-beta.17
+  - @mastra/schema-compat@1.0.0-beta.4
+
+## 1.0.0-beta.8
+
+### Patch Changes
+
+- Add storage composition to MastraStorage ([#11401](https://github.com/mastra-ai/mastra/pull/11401))
+
+  `MastraStorage` can now compose storage domains from different adapters. Use it when you need different databases for different purposes - for example, PostgreSQL for memory and workflows, but a different database for observability.
+
+  ```typescript
+  import { MastraStorage } from '@mastra/core/storage';
+  import { MemoryPG, WorkflowsPG, ScoresPG } from '@mastra/pg';
+  import { MemoryLibSQL } from '@mastra/libsql';
+
+  // Compose domains from different stores
+  const storage = new MastraStorage({
+    id: 'composite',
+    domains: {
+      memory: new MemoryLibSQL({ url: 'file:./local.db' }),
+      workflows: new WorkflowsPG({ connectionString: process.env.DATABASE_URL }),
+      scores: new ScoresPG({ connectionString: process.env.DATABASE_URL }),
+    },
+  });
+  ```
+
+  **Breaking changes:**
+  - `storage.supports` property no longer exists
+  - `StorageSupports` type is no longer exported from `@mastra/core/storage`
+
+  All stores now support the same features. For domain availability, use `getStore()`:
+
+  ```typescript
+  const store = await storage.getStore('memory');
+  if (store) {
+    // domain is available
+  }
+  ```
+
+- Updated dependencies [[`3d93a15`](https://github.com/mastra-ai/mastra/commit/3d93a15796b158c617461c8b98bede476ebb43e2), [`efe406a`](https://github.com/mastra-ai/mastra/commit/efe406a1353c24993280ebc2ed61dd9f65b84b26), [`119e5c6`](https://github.com/mastra-ai/mastra/commit/119e5c65008f3e5cfca954eefc2eb85e3bf40da4), [`74e504a`](https://github.com/mastra-ai/mastra/commit/74e504a3b584eafd2f198001c6a113bbec589fd3), [`e33fdbd`](https://github.com/mastra-ai/mastra/commit/e33fdbd07b33920d81e823122331b0c0bee0bb59), [`929f69c`](https://github.com/mastra-ai/mastra/commit/929f69c3436fa20dd0f0e2f7ebe8270bd82a1529), [`8a73529`](https://github.com/mastra-ai/mastra/commit/8a73529ca01187f604b1f3019d0a725ac63ae55f)]:
+  - @mastra/core@1.0.0-beta.16
+
+## 1.0.0-beta.7
+
+### Patch Changes
+
+- Consolidate memory integration tests and fix working memory filtering in MessageHistory processor ([#11367](https://github.com/mastra-ai/mastra/pull/11367))
+
+  Moved `extractWorkingMemoryTags`, `removeWorkingMemoryTags`, and `extractWorkingMemoryContent` utilities from `@mastra/memory` to `@mastra/core/memory` so they can be used by the `MessageHistory` processor.
+
+  Updated `MessageHistory.filterMessagesForPersistence()` to properly filter out `updateWorkingMemory` tool invocations and strip working memory tags from text content, fixing an issue where working memory tool call arguments were polluting saved message history for v5+ models.
+
+  Also consolidated integration tests for agent-memory, working-memory, and pg-storage into shared test functions that can run against multiple model versions (v4, v5, v6).
+
+- Added support for AI SDK v6 embedding models (specification version v3) in memory and vector modules. Fixed TypeScript error where `ModelRouterEmbeddingModel` was trying to implement a union type instead of `EmbeddingModelV2` directly. ([#11362](https://github.com/mastra-ai/mastra/pull/11362))
+
+- Fix giant debug log messages (30-40kb) when using agents with memory and debug logging enabled. ([#11280](https://github.com/mastra-ai/mastra/pull/11280))
+
+  **Problem**: The memory recall debug log was including the entire threadConfig object, which could contain large zod schemas (30-40kb when serialized).
+
+  **Solution**: Replace the full threadConfig log with selective configuration properties:
+  - `threadId`: Basic identifier for the operation
+  - `perPage`: Pagination setting
+  - `page`: Current page
+  - `orderBy`: Sort configuration
+  - `hasWorkingMemorySchema`: Boolean indicator if schema exists (without the schema itself)
+  - `workingMemoryEnabled`: Working memory enabled status
+  - `semanticRecallEnabled`: Semantic recall enabled status
+
+  This maintains useful debugging information while eliminating the giant log messages containing serialized schemas.
+
+- Introduce StorageDomain base class for composite storage support ([#11249](https://github.com/mastra-ai/mastra/pull/11249))
+
+  Storage adapters now use a domain-based architecture where each domain (memory, workflows, scores, observability, agents) extends a `StorageDomain` base class with `init()` and `dangerouslyClearAll()` methods.
+
+  **Key changes:**
+  - Add `StorageDomain` abstract base class that all domain storage classes extend
+  - Add `InMemoryDB` class for shared state across in-memory domain implementations
+  - All storage domains now implement `dangerouslyClearAll()` for test cleanup
+  - Remove `operations` from public `StorageDomains` type (now internal to each adapter)
+  - Add flexible client/config patterns - domains accept either an existing database client or config to create one internally
+
+  **Why this matters:**
+
+  This enables composite storage where you can use different database adapters per domain:
+
+  ```typescript
+  import { Mastra } from '@mastra/core';
+  import { PostgresStore } from '@mastra/pg';
+  import { ClickhouseStore } from '@mastra/clickhouse';
+
+  // Use Postgres for most domains but Clickhouse for observability
+  const mastra = new Mastra({
+    storage: new PostgresStore({
+      connectionString: 'postgres://...',
+    }),
+    // Future: override specific domains
+    // observability: new ClickhouseStore({ ... }).getStore('observability'),
+  });
+  ```
+
+  **Standalone domain usage:**
+
+  Domains can now be used independently with flexible configuration:
+
+  ```typescript
+  import { MemoryLibSQL } from '@mastra/libsql/memory';
+
+  // Option 1: Pass config to create client internally
+  const memory = new MemoryLibSQL({
+    url: 'file:./local.db',
+  });
+
+  // Option 2: Pass existing client for shared connections
+  import { createClient } from '@libsql/client';
+  const client = createClient({ url: 'file:./local.db' });
+  const memory = new MemoryLibSQL({ client });
+  ```
+
+  **Breaking changes:**
+  - `StorageDomains` type no longer includes `operations` - access via `getStore()` instead
+  - Domain base classes now require implementing `dangerouslyClearAll()` method
+
+- Refactor storage architecture to use domain-specific stores via `getStore()` pattern ([#11361](https://github.com/mastra-ai/mastra/pull/11361))
+
+  ### Summary
+
+  This release introduces a new storage architecture that replaces passthrough methods on `MastraStorage` with domain-specific storage interfaces accessed via `getStore()`. This change reduces code duplication across storage adapters and provides a cleaner, more modular API.
+
+  ### Migration Guide
+
+  All direct method calls on storage instances should be updated to use `getStore()`:
+
+  ```typescript
+  // Before
+  const thread = await storage.getThreadById({ threadId });
+  await storage.persistWorkflowSnapshot({ workflowName, runId, snapshot });
+  await storage.createSpan(span);
+
+  // After
+  const memory = await storage.getStore('memory');
+  const thread = await memory?.getThreadById({ threadId });
+
+  const workflows = await storage.getStore('workflows');
+  await workflows?.persistWorkflowSnapshot({ workflowName, runId, snapshot });
+
+  const observability = await storage.getStore('observability');
+  await observability?.createSpan(span);
+  ```
+
+  ### Available Domains
+  - **`memory`**: Thread and message operations (`getThreadById`, `saveThread`, `saveMessages`, etc.)
+  - **`workflows`**: Workflow state persistence (`persistWorkflowSnapshot`, `loadWorkflowSnapshot`, `getWorkflowRunById`, etc.)
+  - **`scores`**: Evaluation scores (`saveScore`, `listScoresByScorerId`, etc.)
+  - **`observability`**: Tracing and spans (`createSpan`, `updateSpan`, `getTrace`, etc.)
+  - **`agents`**: Stored agent configurations (`createAgent`, `getAgentById`, `listAgents`, etc.)
+
+  ### Breaking Changes
+  - Passthrough methods have been removed from `MastraStorage` base class
+  - All storage adapters now require accessing domains via `getStore()`
+  - The `stores` property on storage instances is now the canonical way to access domain storage
+
+  ### Internal Changes
+  - Each storage adapter now initializes domain-specific stores in its constructor
+  - Domain stores share database connections and handle their own table initialization
+
+- Updated dependencies [[`33a4d2e`](https://github.com/mastra-ai/mastra/commit/33a4d2e4ed8af51f69256232f00c34d6b6b51d48), [`4aaa844`](https://github.com/mastra-ai/mastra/commit/4aaa844a4f19d054490f43638a990cc57bda8d2f), [`4a1a6cb`](https://github.com/mastra-ai/mastra/commit/4a1a6cb3facad54b2bb6780b00ce91d6de1edc08), [`31d13d5`](https://github.com/mastra-ai/mastra/commit/31d13d5fdc2e2380e2e3ee3ec9fb29d2a00f265d), [`4c62166`](https://github.com/mastra-ai/mastra/commit/4c621669f4a29b1f443eca3ba70b814afa286266), [`7bcbf10`](https://github.com/mastra-ai/mastra/commit/7bcbf10133516e03df964b941f9a34e9e4ab4177), [`4353600`](https://github.com/mastra-ai/mastra/commit/43536005a65988a8eede236f69122e7f5a284ba2), [`6986fb0`](https://github.com/mastra-ai/mastra/commit/6986fb064f5db6ecc24aa655e1d26529087b43b3), [`053e979`](https://github.com/mastra-ai/mastra/commit/053e9793b28e970086b0507f7f3b76ea32c1e838), [`e26dc9c`](https://github.com/mastra-ai/mastra/commit/e26dc9c3ccfec54ae3dc3e2b2589f741f9ae60a6), [`55edf73`](https://github.com/mastra-ai/mastra/commit/55edf7302149d6c964fbb7908b43babfc2b52145), [`27c0009`](https://github.com/mastra-ai/mastra/commit/27c0009777a6073d7631b0eb7b481d94e165b5ca), [`dee388d`](https://github.com/mastra-ai/mastra/commit/dee388dde02f2e63c53385ae69252a47ab6825cc), [`3f3fc30`](https://github.com/mastra-ai/mastra/commit/3f3fc3096f24c4a26cffeecfe73085928f72aa63), [`d90ea65`](https://github.com/mastra-ai/mastra/commit/d90ea6536f7aa51c6545a4e9215b55858e98e16d), [`d171e55`](https://github.com/mastra-ai/mastra/commit/d171e559ead9f52ec728d424844c8f7b164c4510), [`10c2735`](https://github.com/mastra-ai/mastra/commit/10c27355edfdad1ee2b826b897df74125eb81fb8), [`1924cf0`](https://github.com/mastra-ai/mastra/commit/1924cf06816e5e4d4d5333065ec0f4bb02a97799), [`b339816`](https://github.com/mastra-ai/mastra/commit/b339816df0984d0243d944ac2655d6ba5f809cde)]:
+  - @mastra/core@1.0.0-beta.15
+
+## 1.0.0-beta.6
+
+### Patch Changes
+
+- Fixed ReDoS vulnerability in working memory tag parsing. ([#11248](https://github.com/mastra-ai/mastra/pull/11248))
+
+  Replaced regex-based parsing with indexOf-based string parsing to prevent denial of service attacks from malicious input. The vulnerable regex `/<working_memory>([^]*?)<\/working_memory>/g` had O(n²) complexity on pathological inputs - the new implementation maintains O(n) linear time.
+
+- Updated dependencies [[`4f94ed8`](https://github.com/mastra-ai/mastra/commit/4f94ed8177abfde3ec536e3574883e075423350c), [`ac3cc23`](https://github.com/mastra-ai/mastra/commit/ac3cc2397d1966bc0fc2736a223abc449d3c7719), [`a86f4df`](https://github.com/mastra-ai/mastra/commit/a86f4df0407311e0d2ea49b9a541f0938810d6a9), [`029540c`](https://github.com/mastra-ai/mastra/commit/029540ca1e582fc2dd8d288ecd4a9b0f31a954ef), [`66741d1`](https://github.com/mastra-ai/mastra/commit/66741d1a99c4f42cf23a16109939e8348ac6852e), [`01b20fe`](https://github.com/mastra-ai/mastra/commit/01b20fefb7c67c2b7d79417598ef4e60256d1225), [`0dbf199`](https://github.com/mastra-ai/mastra/commit/0dbf199110f22192ce5c95b1c8148d4872b4d119), [`a7ce182`](https://github.com/mastra-ai/mastra/commit/a7ce1822a8785ce45d62dd5c911af465e144f7d7)]:
+  - @mastra/core@1.0.0-beta.14
+
+## 1.0.0-beta.5
+
+### Patch Changes
+
+- Embed AI types to fix peerdeps mismatches ([`9650cce`](https://github.com/mastra-ai/mastra/commit/9650cce52a1d917ff9114653398e2a0f5c3ba808))
+
+- Fix crash in updateMessageToHideWorkingMemoryV2 when message.content is not a V2 object. Added defensive type guards before spreading content to handle legacy or malformed message formats. ([#11180](https://github.com/mastra-ai/mastra/pull/11180))
+
+- Updated dependencies [[`d5ed981`](https://github.com/mastra-ai/mastra/commit/d5ed981c8701c1b8a27a5f35a9a2f7d9244e695f), [`9650cce`](https://github.com/mastra-ai/mastra/commit/9650cce52a1d917ff9114653398e2a0f5c3ba808), [`932d63d`](https://github.com/mastra-ai/mastra/commit/932d63dd51be9c8bf1e00e3671fe65606c6fb9cd), [`b760b73`](https://github.com/mastra-ai/mastra/commit/b760b731aca7c8a3f041f61d57a7f125ae9cb215), [`695a621`](https://github.com/mastra-ai/mastra/commit/695a621528bdabeb87f83c2277cf2bb084c7f2b4), [`2b459f4`](https://github.com/mastra-ai/mastra/commit/2b459f466fd91688eeb2a44801dc23f7f8a887ab), [`486352b`](https://github.com/mastra-ai/mastra/commit/486352b66c746602b68a95839f830de14c7fb8c0), [`09e4bae`](https://github.com/mastra-ai/mastra/commit/09e4bae18dd5357d2ae078a4a95a2af32168ab08), [`24b76d8`](https://github.com/mastra-ai/mastra/commit/24b76d8e17656269c8ed09a0c038adb9cc2ae95a), [`243a823`](https://github.com/mastra-ai/mastra/commit/243a8239c5906f5c94e4f78b54676793f7510ae3), [`486352b`](https://github.com/mastra-ai/mastra/commit/486352b66c746602b68a95839f830de14c7fb8c0), [`c61fac3`](https://github.com/mastra-ai/mastra/commit/c61fac3add96f0dcce0208c07415279e2537eb62), [`6f14f70`](https://github.com/mastra-ai/mastra/commit/6f14f706ccaaf81b69544b6c1b75ab66a41e5317), [`09e4bae`](https://github.com/mastra-ai/mastra/commit/09e4bae18dd5357d2ae078a4a95a2af32168ab08), [`4524734`](https://github.com/mastra-ai/mastra/commit/45247343e384717a7c8404296275c56201d6470f), [`2a53598`](https://github.com/mastra-ai/mastra/commit/2a53598c6d8cfeb904a7fc74e57e526d751c8fa6), [`c7cd3c7`](https://github.com/mastra-ai/mastra/commit/c7cd3c7a187d7aaf79e2ca139de328bf609a14b4), [`847c212`](https://github.com/mastra-ai/mastra/commit/847c212caba7df0d6f2fc756b494ac3c75c3720d), [`6f941c4`](https://github.com/mastra-ai/mastra/commit/6f941c438ca5f578619788acc7608fc2e23bd176)]:
+  - @mastra/core@1.0.0-beta.12
+  - @mastra/schema-compat@1.0.0-beta.3
+
+## 1.0.0-beta.4
+
+### Patch Changes
+
+- Fix connection pool exhaustion when saving many messages with semantic recall enabled. Instead of calling vector.upsert() for each message individually (which acquires a separate DB connection), all embeddings are now batched into a single upsert call. ([#10954](https://github.com/mastra-ai/mastra/pull/10954))
+
+- Updated dependencies [[`0d41fe2`](https://github.com/mastra-ai/mastra/commit/0d41fe245355dfc66d61a0d9c85d9400aac351ff), [`6b3ba91`](https://github.com/mastra-ai/mastra/commit/6b3ba91494cc10394df96782f349a4f7b1e152cc), [`7907fd1`](https://github.com/mastra-ai/mastra/commit/7907fd1c5059813b7b870b81ca71041dc807331b)]:
+  - @mastra/core@1.0.0-beta.8
+
+## 1.0.0-beta.3
+
+### Minor Changes
+
+- Memory system now uses processors. Memory processors (`MessageHistory`, `SemanticRecall`, `WorkingMemory`) are now exported from `@mastra/memory/processors` and automatically added to the agent pipeline based on your memory config. Core processors (`ToolCallFilter`, `TokenLimiter`) remain in `@mastra/core/processors`. ([#9254](https://github.com/mastra-ai/mastra/pull/9254))
+
+- Schema-based working memory now uses merge semantics instead of replace semantics. ([#10659](https://github.com/mastra-ai/mastra/pull/10659))
+
+  **Before:** Each working memory update replaced the entire memory, causing data loss across conversation turns.
+
+  **After:** For schema-based working memory:
+  - Object fields are deep merged (existing fields preserved, only provided fields updated)
+  - Set a field to `null` to delete it
+  - Arrays are replaced entirely when provided
+
+  Template-based (Markdown) working memory retains the existing replace semantics.
+
+  This fixes issue #7775 where users building profile-like schemas would lose information from previous turns.
+
+### Patch Changes
+
+- feat(storage): support querying messages from multiple threads ([#10663](https://github.com/mastra-ai/mastra/pull/10663))
+  - Fixed TypeScript errors where `threadId: string | string[]` was being passed to places expecting `Scalar` type
+  - Added proper multi-thread support for `listMessages` across all adapters when `threadId` is an array
+  - Updated `_getIncludedMessages` to look up message threadId by ID (since message IDs are globally unique)
+  - **upstash**: Added `msg-idx:{messageId}` index for O(1) message lookups (backwards compatible with fallback to scan for old messages, with automatic backfill)
+
+- Fix recall() to return newest messages when using lastMessages config ([#10543](https://github.com/mastra-ai/mastra/pull/10543))
+
+  When using lastMessages: N config without an explicit orderBy, the recall() function was returning the OLDEST N messages instead of the NEWEST N messages. This completely breaks conversation history for any thread that grows beyond the lastMessages limit.
+
+- Updated dependencies [[`ac0d2f4`](https://github.com/mastra-ai/mastra/commit/ac0d2f4ff8831f72c1c66c2be809706d17f65789), [`1a0d3fc`](https://github.com/mastra-ai/mastra/commit/1a0d3fc811482c9c376cdf79ee615c23bae9b2d6), [`85a628b`](https://github.com/mastra-ai/mastra/commit/85a628b1224a8f64cd82ea7f033774bf22df7a7e), [`c237233`](https://github.com/mastra-ai/mastra/commit/c23723399ccedf7f5744b3f40997b79246bfbe64), [`15f9e21`](https://github.com/mastra-ai/mastra/commit/15f9e216177201ea6e3f6d0bfb063fcc0953444f), [`ff94dea`](https://github.com/mastra-ai/mastra/commit/ff94dea935f4e34545c63bcb6c29804732698809), [`5b2ff46`](https://github.com/mastra-ai/mastra/commit/5b2ff4651df70c146523a7fca773f8eb0a2272f8), [`db41688`](https://github.com/mastra-ai/mastra/commit/db4168806d007417e2e60b4f68656dca4e5f40c9), [`5ca599d`](https://github.com/mastra-ai/mastra/commit/5ca599d0bb59a1595f19f58473fcd67cc71cef58), [`bff1145`](https://github.com/mastra-ai/mastra/commit/bff114556b3cbadad9b2768488708f8ad0e91475), [`5c8ca24`](https://github.com/mastra-ai/mastra/commit/5c8ca247094e0cc2cdbd7137822fb47241f86e77), [`e191844`](https://github.com/mastra-ai/mastra/commit/e1918444ca3f80e82feef1dad506cd4ec6e2875f), [`22553f1`](https://github.com/mastra-ai/mastra/commit/22553f11c63ee5e966a9c034a349822249584691), [`7237163`](https://github.com/mastra-ai/mastra/commit/72371635dbf96a87df4b073cc48fc655afbdce3d), [`2500740`](https://github.com/mastra-ai/mastra/commit/2500740ea23da067d6e50ec71c625ab3ce275e64), [`873ecbb`](https://github.com/mastra-ai/mastra/commit/873ecbb517586aa17d2f1e99283755b3ebb2863f), [`4f9bbe5`](https://github.com/mastra-ai/mastra/commit/4f9bbe5968f42c86f4930b8193de3c3c17e5bd36), [`02e51fe`](https://github.com/mastra-ai/mastra/commit/02e51feddb3d4155cfbcc42624fd0d0970d032c0), [`8f3fa3a`](https://github.com/mastra-ai/mastra/commit/8f3fa3a652bb77da092f913ec51ae46e3a7e27dc), [`cd29ad2`](https://github.com/mastra-ai/mastra/commit/cd29ad23a255534e8191f249593849ed29160886), [`bdf4d8c`](https://github.com/mastra-ai/mastra/commit/bdf4d8cdc656d8a2c21d81834bfa3bfa70f56c16), [`854e3da`](https://github.com/mastra-ai/mastra/commit/854e3dad5daac17a91a20986399d3a51f54bf68b), [`ce18d38`](https://github.com/mastra-ai/mastra/commit/ce18d38678c65870350d123955014a8432075fd9), [`cccf9c8`](https://github.com/mastra-ai/mastra/commit/cccf9c8b2d2dfc1a5e63919395b83d78c89682a0), [`61a5705`](https://github.com/mastra-ai/mastra/commit/61a570551278b6743e64243b3ce7d73de915ca8a), [`db70a48`](https://github.com/mastra-ai/mastra/commit/db70a48aeeeeb8e5f92007e8ede52c364ce15287), [`f0fdc14`](https://github.com/mastra-ai/mastra/commit/f0fdc14ee233d619266b3d2bbdeea7d25cfc6d13), [`db18bc9`](https://github.com/mastra-ai/mastra/commit/db18bc9c3825e2c1a0ad9a183cc9935f6691bfa1), [`9b37b56`](https://github.com/mastra-ai/mastra/commit/9b37b565e1f2a76c24f728945cc740c2b09be9da), [`41a23c3`](https://github.com/mastra-ai/mastra/commit/41a23c32f9877d71810f37e24930515df2ff7a0f), [`5d171ad`](https://github.com/mastra-ai/mastra/commit/5d171ad9ef340387276b77c2bb3e83e83332d729), [`f03ae60`](https://github.com/mastra-ai/mastra/commit/f03ae60500fe350c9d828621006cdafe1975fdd8), [`d1e74a0`](https://github.com/mastra-ai/mastra/commit/d1e74a0a293866dece31022047f5dbab65a304d0), [`39e7869`](https://github.com/mastra-ai/mastra/commit/39e7869bc7d0ee391077ce291474d8a84eedccff), [`5761926`](https://github.com/mastra-ai/mastra/commit/57619260c4a2cdd598763abbacd90de594c6bc76), [`c900fdd`](https://github.com/mastra-ai/mastra/commit/c900fdd504c41348efdffb205cfe80d48c38fa33), [`604a79f`](https://github.com/mastra-ai/mastra/commit/604a79fecf276e26a54a3fe01bb94e65315d2e0e), [`887f0b4`](https://github.com/mastra-ai/mastra/commit/887f0b4746cdbd7cb7d6b17ac9f82aeb58037ea5), [`2562143`](https://github.com/mastra-ai/mastra/commit/256214336b4faa78646c9c1776612393790d8784), [`ef11a61`](https://github.com/mastra-ai/mastra/commit/ef11a61920fa0ed08a5b7ceedd192875af119749)]:
+  - @mastra/core@1.0.0-beta.6
+  - @mastra/schema-compat@1.0.0-beta.2
+
+## 1.0.0-beta.2
+
+### Patch Changes
+
+- Add new deleteVectors, updateVector by filter ([#10408](https://github.com/mastra-ai/mastra/pull/10408))
+
+- Update MockMemory to work with new storage API changes. MockMemory now properly implements all abstract MastraMemory methods. This includes proper thread management, message saving with MessageList conversion, working memory operations with scope support, and resource listing. ([#10368](https://github.com/mastra-ai/mastra/pull/10368))
+
+  Add Zod v4 support for working memory schemas. Memory implementations now check for Zod v4's built-in `.toJsonSchema()` method before falling back to the `zodToJsonSchema` compatibility function, improving performance and forward compatibility while maintaining backward compatibility with Zod v3.
+
+  Add Gemini 3 Pro test coverage in agent-gemini.test.ts to validate the latest Gemini model integration.
+
+- Fix working memory zod to json schema conversion to use schema-compat zodtoJsonSchema fn. ([#10391](https://github.com/mastra-ai/mastra/pull/10391))
+
+- Updated dependencies [[`21a15de`](https://github.com/mastra-ai/mastra/commit/21a15de369fe82aac26bb642ed7be73505475e8b), [`feb7ee4`](https://github.com/mastra-ai/mastra/commit/feb7ee4d09a75edb46c6669a3beaceec78811747), [`b0e2ea5`](https://github.com/mastra-ai/mastra/commit/b0e2ea5b52c40fae438b9e2f7baee6f0f89c5442), [`c456e01`](https://github.com/mastra-ai/mastra/commit/c456e0149e3c176afcefdbd9bb1d2c5917723725), [`ab035c2`](https://github.com/mastra-ai/mastra/commit/ab035c2ef6d8cc7bb25f06f1a38508bd9e6f126b), [`1a46a56`](https://github.com/mastra-ai/mastra/commit/1a46a566f45a3fcbadc1cf36bf86d351f264bfa3), [`3cf540b`](https://github.com/mastra-ai/mastra/commit/3cf540b9fbfea8f4fc8d3a2319a4e6c0b0cbfd52), [`1c6ce51`](https://github.com/mastra-ai/mastra/commit/1c6ce51f875915ab57fd36873623013699a2a65d), [`898a972`](https://github.com/mastra-ai/mastra/commit/898a9727d286c2510d6b702dfd367e6aaf5c6b0f), [`a97003a`](https://github.com/mastra-ai/mastra/commit/a97003aa1cf2f4022a41912324a1e77263b326b8), [`ccc141e`](https://github.com/mastra-ai/mastra/commit/ccc141ed27da0abc3a3fc28e9e5128152e8e37f4), [`fe3b897`](https://github.com/mastra-ai/mastra/commit/fe3b897c2ccbcd2b10e81b099438c7337feddf89), [`00123ba`](https://github.com/mastra-ai/mastra/commit/00123ba96dc9e5cd0b110420ebdba56d8f237b25), [`29c4309`](https://github.com/mastra-ai/mastra/commit/29c4309f818b24304c041bcb4a8f19b5f13f6b62), [`16785ce`](https://github.com/mastra-ai/mastra/commit/16785ced928f6f22638f4488cf8a125d99211799), [`de8239b`](https://github.com/mastra-ai/mastra/commit/de8239bdcb1d8c0cfa06da21f1569912a66bbc8a), [`b5e6cd7`](https://github.com/mastra-ai/mastra/commit/b5e6cd77fc8c8e64e0494c1d06cee3d84e795d1e), [`3759cb0`](https://github.com/mastra-ai/mastra/commit/3759cb064935b5f74c65ac2f52a1145f7352899d), [`651e772`](https://github.com/mastra-ai/mastra/commit/651e772eb1475fb13e126d3fcc01751297a88214), [`b61b93f`](https://github.com/mastra-ai/mastra/commit/b61b93f9e058b11dd2eec169853175d31dbdd567), [`bae33d9`](https://github.com/mastra-ai/mastra/commit/bae33d91a63fbb64d1e80519e1fc1acaed1e9013), [`c0b731f`](https://github.com/mastra-ai/mastra/commit/c0b731fb27d712dc8582e846df5c0332a6a0c5ba), [`43ca8f2`](https://github.com/mastra-ai/mastra/commit/43ca8f2c7334851cc7b4d3d2f037d8784bfbdd5f), [`2ca67cc`](https://github.com/mastra-ai/mastra/commit/2ca67cc3bb1f6a617353fdcab197d9efebe60d6f), [`9e67002`](https://github.com/mastra-ai/mastra/commit/9e67002b52c9be19936c420a489dbee9c5fd6a78), [`35edc49`](https://github.com/mastra-ai/mastra/commit/35edc49ac0556db609189641d6341e76771b81fc)]:
+  - @mastra/core@1.0.0-beta.5
+  - @mastra/schema-compat@1.0.0-beta.1
+
 ## 1.0.0-beta.1
 
 ### Patch Changes