@mastra/mcp-docs-server 0.13.0 → 0.13.1-alpha.2

package/.docs/organized/changelogs/mastra.md

@@ -1,5 +1,128 @@
 # mastra
 
+## 0.10.6-alpha.7
+
+### Patch Changes
+
+- b29c802: Remove open browser
+- Updated dependencies [bd1674f]
+- Updated dependencies [69f76f7]
+  - @mastra/mcp@0.10.4-alpha.1
+  - @mastra/deployer@0.10.6-alpha.5
+
+## 0.10.6-alpha.6
+
+### Patch Changes
+
+- 5f2aa3e: Move workflow hooks to playground
+- Updated dependencies [12a95fc]
+- Updated dependencies [51264a5]
+- Updated dependencies [8e6f677]
+  - @mastra/core@0.10.6-alpha.5
+  - @mastra/deployer@0.10.6-alpha.5
+
+## 0.10.6-alpha.5
+
+### Patch Changes
+
+- 084f6aa: Add logs to circular dependency to warn people when starting server might break
+- Updated dependencies [79b9909]
+- Updated dependencies [084f6aa]
+- Updated dependencies [9589624]
+  - @mastra/deployer@0.10.6-alpha.4
+  - @mastra/core@0.10.6-alpha.4
+
+## 0.10.6-alpha.4
+
+### Patch Changes
+
+- 4051477: dependencies updates:
+  - Updated dependency [`@clack/prompts@^0.11.0` ↗︎](https://www.npmjs.com/package/@clack/prompts/v/0.11.0) (from `^0.8.2`, in `dependencies`)
+  - Updated dependency [`@opentelemetry/instrumentation@^0.202.0` ↗︎](https://www.npmjs.com/package/@opentelemetry/instrumentation/v/0.202.0) (from `^0.57.2`, in `dependencies`)
+  - Updated dependency [`posthog-node@4.18.0` ↗︎](https://www.npmjs.com/package/posthog-node/v/4.18.0) (from `4.16.0`, in `dependencies`)
+  - Updated dependency [`yocto-spinner@^0.2.3` ↗︎](https://www.npmjs.com/package/yocto-spinner/v/0.2.3) (from `^0.1.2`, in `dependencies`)
+  - Added dependency [`open@^10.1.2` ↗︎](https://www.npmjs.com/package/open/v/10.1.2) (to `dependencies`)
+- b40f365: dependencies updates:
+  - Updated dependency [`@clack/prompts@^0.11.0` ↗︎](https://www.npmjs.com/package/@clack/prompts/v/0.11.0) (from `^0.8.2`, in `dependencies`)
+  - Updated dependency [`@opentelemetry/instrumentation@^0.202.0` ↗︎](https://www.npmjs.com/package/@opentelemetry/instrumentation/v/0.202.0) (from `^0.57.2`, in `dependencies`)
+  - Updated dependency [`posthog-node@4.18.0` ↗︎](https://www.npmjs.com/package/posthog-node/v/4.18.0) (from `4.16.0`, in `dependencies`)
+  - Updated dependency [`yocto-spinner@^0.2.3` ↗︎](https://www.npmjs.com/package/yocto-spinner/v/0.2.3) (from `^0.1.2`, in `dependencies`)
+- Updated dependencies [4051477]
+- Updated dependencies [c28ed65]
+- Updated dependencies [d70c420]
+- Updated dependencies [2a16996]
+  - @mastra/deployer@0.10.6-alpha.3
+  - @mastra/core@0.10.6-alpha.3
+
+## 0.10.6-alpha.3
+
+### Patch Changes
+
+- Updated dependencies [4b0f8a6]
+- Updated dependencies [ec7f824]
+  - @mastra/core@0.10.6-alpha.2
+  - @mastra/deployer@0.10.6-alpha.2
+
+## 0.10.6-alpha.2
+
+### Patch Changes
+
+- 143b4e4: Fix globbing of tools to only capture js/ts files
+- 44ba52d: Add proper error message when installation of mastra fails
+- 3270d9d: Fix runtime context being undefined
+- fc677d7: For final result for a workflow
+- Updated dependencies [ee9af57]
+- Updated dependencies [3270d9d]
+- Updated dependencies [751c894]
+- Updated dependencies [577ce3a]
+- Updated dependencies [9260b3a]
+  - @mastra/deployer@0.10.6-alpha.1
+  - @mastra/core@0.10.6-alpha.1
+
+## 0.10.6-alpha.1
+
+### Patch Changes
+
+- 47e7029: Add open browser functionality when running mastra dev
+
+## 0.10.6-alpha.0
+
+### Patch Changes
+
+- 02560d4: lift evals fetching to the playground package instead
+- 63f6b7d: dependencies updates:
+  - Updated dependency [`execa@^9.6.0` ↗︎](https://www.npmjs.com/package/execa/v/9.6.0) (from `^9.5.2`, in `dependencies`)
+  - Updated dependency [`json-schema-to-zod@^2.6.1` ↗︎](https://www.npmjs.com/package/json-schema-to-zod/v/2.6.1) (from `^2.6.0`, in `dependencies`)
+  - Updated dependency [`shell-quote@^1.8.3` ↗︎](https://www.npmjs.com/package/shell-quote/v/1.8.3) (from `^1.8.2`, in `dependencies`)
+  - Updated dependency [`strip-json-comments@^5.0.2` ↗︎](https://www.npmjs.com/package/strip-json-comments/v/5.0.2) (from `^5.0.1`, in `dependencies`)
+  - Updated dependency [`zod@^3.25.57` ↗︎](https://www.npmjs.com/package/zod/v/3.25.57) (from `^3.25.56`, in `dependencies`)
+- 311132e: move useWorkflow to playground instead of playground-ui
+- 906f992: CLI error log
+- 53d3c37: Get workflows from an agent if not found from Mastra instance #5083
+- Updated dependencies [63f6b7d]
+- Updated dependencies [2d12edd]
+- Updated dependencies [63f6b7d]
+- Updated dependencies [63f6b7d]
+- Updated dependencies [36f1c36]
+- Updated dependencies [10d352e]
+- Updated dependencies [53d3c37]
+  - @mastra/core@0.10.6-alpha.0
+  - @mastra/deployer@0.10.6-alpha.0
+  - @mastra/mcp@0.10.4-alpha.0
+
+## 0.10.5
+
+### Patch Changes
+
+- 1ba421d: fix the tools not showing on workflows attached to agents
+- 8725d02: Improve cli by reducing the amount of setups during interactive prompt
+- 13c97f9: Save run status, result and error in storage snapshot
+- Updated dependencies [8725d02]
+- Updated dependencies [13c97f9]
+- Updated dependencies [105f872]
+  - @mastra/deployer@0.10.5
+  - @mastra/core@0.10.5
+
 ## 0.10.4
 
 ### Patch Changes
@@ -175,128 +298,5 @@
 - 6fd77b5: add docs and txt support for multi modal
 - 9faee5b: small fixes in the workflows graph
 - 631683f: move workflow runs list in playground-ui instead of playground
-- 068b850: fix: able to pass headers to playground components which are using the mastra client
-- 668af6d: Fix mastra dev server restarts
-- f0d559f: Fix peerdeps for alpha channel
-- f6ddf55: fix traces not showing and reduce API surface from playground ui
-- 9a31c09: Highlight steps in nested workflows on workflow graph
-- Updated dependencies [ee77e78]
-- Updated dependencies [592a2db]
-- Updated dependencies [e5dc18d]
-- Updated dependencies [ab5adbe]
-- Updated dependencies [e8d2aff]
-- Updated dependencies [1e8bb40]
-- Updated dependencies [1b5fc55]
-- Updated dependencies [195c428]
-- Updated dependencies [f73e11b]
-- Updated dependencies [37643b8]
-- Updated dependencies [99fd6cf]
-- Updated dependencies [1fcc048]
-- Updated dependencies [c5bf1ce]
-- Updated dependencies [f946acf]
-- Updated dependencies [add596e]
-- Updated dependencies [8dc94d8]
-- Updated dependencies [ecebbeb]
-- Updated dependencies [4187ed4]
-- Updated dependencies [79d5145]
-- Updated dependencies [12b7002]
-- Updated dependencies [f0d559f]
-- Updated dependencies [2901125]
-  - @mastra/core@0.10.2
-  - @mastra/mcp@0.10.2
-  - @mastra/deployer@0.10.2
-  - @mastra/loggers@0.10.1
-
-## 0.10.2-alpha.9
-
-### Patch Changes
-
-- 90e96de: Fix: prevent default flag from triggering interactive prompt
-- Updated dependencies [37643b8]
-- Updated dependencies [79d5145]
-  - @mastra/core@0.10.2-alpha.8
-  - @mastra/deployer@0.10.2-alpha.8
-
-## 0.10.2-alpha.8
-
-### Patch Changes
-
-- a399086: Bumping because we forgot to
-  - @mastra/deployer@0.10.2-alpha.7
-  - @mastra/core@0.10.2-alpha.7
-
-## 0.10.2-alpha.7
-
-### Patch Changes
-
-- 1fcc048: chore: generate sourcemaps in dev build
-- 6fd77b5: add docs and txt support for multi modal
-- 631683f: move workflow runs list in playground-ui instead of playground
-- Updated dependencies [99fd6cf]
-- Updated dependencies [1fcc048]
-- Updated dependencies [8dc94d8]
-  - @mastra/core@0.10.2-alpha.6
-  - @mastra/deployer@0.10.2-alpha.6
-
-## 0.10.2-alpha.6
-
-### Patch Changes
-
-- 1b5fc55: Fixed an issue where the playground wouldn't display images saved in memory. Fixed memory to always store images as strings. Removed duplicate storage of reasoning and file/image parts in storage dbs
-- 9666468: move the fetch traces call to the playground instead of playground-ui
-- 668af6d: Fix mastra dev server restarts
-- Updated dependencies [1b5fc55]
-- Updated dependencies [add596e]
-- Updated dependencies [ecebbeb]
-  - @mastra/core@0.10.2-alpha.5
-  - @mastra/deployer@0.10.2-alpha.5
-
-## 0.10.2-alpha.5
-
-### Patch Changes
-
-- 401bbae: Show workflow graph from stepGraph of previous runs when viewing a previous run
-
-## 0.10.2-alpha.4
-
-### Patch Changes
-
-- Updated dependencies [c5bf1ce]
-- Updated dependencies [12b7002]
-  - @mastra/core@0.10.2-alpha.4
-  - @mastra/deployer@0.10.2-alpha.4
-
-## 0.10.2-alpha.3
-
-### Patch Changes
-
-- f73e11b: fix telemetry disabled not working on playground
-- 068b850: fix: able to pass headers to playground components which are using the mastra client
-- Updated dependencies [ab5adbe]
-- Updated dependencies [195c428]
-- Updated dependencies [f73e11b]
-- Updated dependencies [f946acf]
-  - @mastra/core@0.10.2-alpha.3
-  - @mastra/deployer@0.10.2-alpha.3
-
-## 0.10.2-alpha.2
-
-### Patch Changes
-
-- 73fec0b: Mastra start cli command"
-- f0d559f: Fix peerdeps for alpha channel
-- f6ddf55: fix traces not showing and reduce API surface from playground ui
-- Updated dependencies [e8d2aff]
-- Updated dependencies [1e8bb40]
-- Updated dependencies [4187ed4]
-- Updated dependencies [f0d559f]
-  - @mastra/deployer@0.10.2-alpha.2
-  - @mastra/core@0.10.2-alpha.2
-  - @mastra/loggers@0.10.1-alpha.0
-  - @mastra/mcp@0.10.2-alpha.1
-
-## 0.10.2-alpha.1
-
-### Patch Changes
 
-... 3711 more lines hidden. See full changelog in package directory.
+... 3834 more lines hidden. See full changelog in package directory.

package/.docs/organized/code-examples/agent.md

@@ -16,7 +16,7 @@
     "@mastra/memory": "latest",
     "@mastra/voice-openai": "latest",
     "@mastra/libsql": "latest",
-    "ai": "^4.3.15",
+    "ai": "^4.3.16",
     "fetch-to-node": "^2.1.0",
     "mastra": "latest",
     "zod": "^3.25.56"
@@ -441,7 +441,7 @@ export const dynamicAgent = new Agent({
 });
 
 export const chefAgentResponses = new Agent({
-  name: 'Chef Agent',
+  name: 'Chef Agent Responses',
   instructions: `
     You are Michel, a practical and experienced home chef who helps people cook great meals with whatever 
     ingredients they have available. Your first priority is understanding what ingredients and equipment the user has access to, then suggesting achievable recipes. 
@@ -451,6 +451,9 @@ export const chefAgentResponses = new Agent({
   tools: {
     web_search_preview: openai.tools.webSearchPreview(),
   },
+  workflows: {
+    myWorkflow,
+  },
 });
 
 ```

package/.docs/organized/code-examples/agui.md

@@ -21,16 +21,19 @@
     "@mastra/agui": "latest",
     "@mastra/core": "latest",
     "@mastra/libsql": "latest",
+    "@mastra/loggers": "latest",
     "@mastra/memory": "latest",
     "mastra": "latest",
     "react": "^19.1.0",
-    "react-dom": "^19.1.0"
+    "react-dom": "^19.1.0",
+    "zod": "^3.25.57"
   },
   "pnpm": {
     "overrides": {
       "@mastra/core": "link:../../packages/core",
       "@mastra/agui": "link:../../packages/agui",
       "@mastra/libsql": "link:../../stores/libsql",
+      "@mastra/loggers": "link:../../packages/loggers",
       "@mastra/memory": "link:../../packages/memory",
       "mastra": "link:../../packages/cli"
     }

package/.docs/organized/code-examples/ai-sdk-useChat.md

@@ -11,7 +11,7 @@
     "lint": "next lint"
   },
   "dependencies": {
-    "@ai-sdk/anthropic": "^1.2.11",
+    "@ai-sdk/anthropic": "^1.2.12",
     "@ai-sdk/react": "^1.2.12",
     "@mastra/core": "latest",
     "next": "^15.3.1",

package/.docs/organized/code-examples/fireworks-r1.md

@@ -13,7 +13,7 @@
   "author": "",
   "license": "MIT",
   "dependencies": {
-    "@ai-sdk/anthropic": "^1.2.11",
+    "@ai-sdk/anthropic": "^1.2.12",
     "@ai-sdk/fireworks": "^0.2.14",
     "@mastra/core": "latest",
     "@mastra/mcp": "latest",

package/.docs/organized/code-examples/memory-with-processors.md

@@ -14,7 +14,7 @@
     "@ai-sdk/openai": "latest",
     "@mastra/core": "latest",
     "@mastra/memory": "latest",
-    "ai": "^4.3.15",
+    "ai": "^4.3.16",
     "chalk": "^5.4.1",
     "dotenv": "^16.3.1",
     "js-tiktoken": "^1.0.13",
@@ -249,7 +249,7 @@ const interviewMemory = new Memory({
 export const interviewerAgent = new Agent({
   name: 'Forgetful Job Interviewer',
   instructions:
-    "You are a professional job interviewer for a technology company. Conduct insightful interviews by asking relevant questions about skills, experience, and problem-solving abilities. Respond to candidate answers and ask follow-up questions. Keep the interview professional and engaging. Remember details the candidate shares earlier in the conversation. Sometimes you forget things by accident. The system will show you if you forgot. Don't be embarassed, you can admit when you forget something, you'll know when you do because there will be a message wrapped in <forgetten> tags. Don't refer to the user by their name, it comes across as too eager",
+    "You are a professional job interviewer for a technology company. Conduct insightful interviews by asking relevant questions about skills, experience, and problem-solving abilities. Respond to candidate answers and ask follow-up questions. Keep the interview professional and engaging. Remember details the candidate shares earlier in the conversation. Sometimes you forget things by accident. The system will show you if you forgot. Don't be embarrassed, you can admit when you forget something, you'll know when you do because there will be a message wrapped in <forgetten> tags. Don't refer to the user by their name, it comes across as too eager",
   model: openai('gpt-4o'),
   memory: interviewMemory,
 });

package/.docs/organized/code-examples/openapi-spec-writer.md

@@ -24,7 +24,7 @@
     "class-variance-authority": "^0.7.1",
     "clsx": "^2.1.1",
     "lucide-react": "^0.454.0",
-    "next": "15.2.3",
+    "next": "15.2.4",
     "react": "^19.0.0",
     "react-dom": "^19.0.0",
     "react-syntax-highlighter": "^15.6.1",

package/.docs/organized/code-examples/quick-start.md

@@ -19,7 +19,7 @@
     "zod": "^3.25.56"
   },
   "dependencies": {
-    "@ai-sdk/anthropic": "^1.1.15",
+    "@ai-sdk/anthropic": "^1.2.12",
     "@ai-sdk/groq": "latest",
     "@ai-sdk/openai": "latest",
     "@mastra/core": "latest",

package/.docs/organized/code-examples/weather-agent.md

@@ -15,12 +15,14 @@
   "dependencies": {
     "@ai-sdk/openai": "^1.3.22",
     "@mastra/core": "latest",
+    "@mastra/libsql": "latest",
     "zod": "^3.25.56"
   },
   "version": "0.0.1",
   "pnpm": {
     "overrides": {
-      "@mastra/core": "link:../../packages/core"
+      "@mastra/core": "link:../../packages/core",
+      "@mastra/libsql": "link:../../stores/libsql"
     }
   },
   "packageManager": "pnpm@10.10.0+sha512.d615db246fe70f25dcfea6d8d73dee782ce23e2245e3c4f6f888249fb568149318637dca73c2c5c8ef2a4ca0d5657fb9567188bfab47f566d1ee6ce987815c39"
@@ -65,12 +67,16 @@ export const weatherReporterAgent = new Agent({
 ### mastra/index.ts
 ```typescript
 import { Mastra } from '@mastra/core';
+import { LibSQLStore } from '@mastra/libsql';
 
 import { weatherAgent } from './agents';
 import { weatherWorkflow as legacyWeatherWorkflow } from './workflows';
 import { weatherWorkflow, weatherWorkflow2 } from './workflows/new-workflow';
 
 export const mastra = new Mastra({
+  storage: new LibSQLStore({
+    url: 'file:./mastra.db',
+  }),
   agents: { weatherAgent },
   legacy_workflows: { legacyWeatherWorkflow },
   workflows: { weatherWorkflow, weatherWorkflow2 },

package/.docs/raw/course/01-first-agent/03-verifying-installation.md

@@ -1,11 +1,13 @@
 # Verifying Your Mastra Installation
 
-Before we begin building our agent, let's make sure you have Mastra properly installed. First, let's check if @mastra/core is in the package.json and there is a src/mastra directory. If so, you can skip this step.
+Before we begin building our agent, let's make sure you have the right development environment set up. Let's check if you have Node.js 18.x or later installed.
+
+Then, let's check if @mastra/core is in the package.json and there is a src/mastra directory. If so, you can skip this step.
 
 If you haven't installed Mastra yet, you can do so by running:
 
 ```bash
-npm create mastra@latest
+npm -y create mastra@latest
 ```
 
 If you do need to install mastra, follow the on-screen prompts and make sure to:

package/.docs/raw/course/01-first-agent/16-adding-memory-to-agent.md

@@ -24,7 +24,7 @@ export const financialAgent = new Agent({
   tools: { getTransactionsTool },
   memory: new Memory({
     storage: new LibSQLStore({
-      url: "../../memory.db", // local file-system database. Location is relative to the output directory `.mastra/output`
+      url: "file:../../memory.db", // local file-system database. Location is relative to the output directory `.mastra/output`
     }),
   }), // Add memory here
 });

package/.docs/raw/course/02-agent-tools-mcp/15-updating-mcp-config-github.md

@@ -3,7 +3,7 @@
 Now, let's update your MCP configuration in `src/mastra/agents/index.ts` to include the GitHub server:
 
 ```typescript
-const mcp = new MCPConfiguration({
+const mcp = new MCPClient({
   servers: {
     zapier: {
       url: new URL(process.env.ZAPIER_MCP_URL || ""),

package/.docs/raw/course/02-agent-tools-mcp/20-updating-mcp-config-hackernews.md

@@ -5,7 +5,7 @@ Unlike the previous MCP servers that use URLs, the Hacker News MCP server can be
 Let's update your MCP configuration in `src/mastra/agents/index.ts` to include the Hacker News server:
 
 ```typescript
-const mcp = new MCPConfiguration({
+const mcp = new MCPClient({
   servers: {
     zapier: {
       url: new URL(process.env.ZAPIER_MCP_URL || ""),

package/.docs/raw/course/02-agent-tools-mcp/26-updating-mcp-config-filesystem.md

@@ -5,7 +5,7 @@ Now, let's update your MCP configuration in `src/mastra/agents/index.ts` to incl
 ```typescript
 import path from "path";
 
-const mcp = new MCPConfiguration({
+const mcp = new MCPClient({
   servers: {
     zapier: {
       url: new URL(process.env.ZAPIER_MCP_URL || ""),

package/.docs/raw/course/03-agent-memory/03-installing-memory.md

@@ -3,9 +3,11 @@
 Let's start by installing the Mastra memory package:
 
 ```bash
-npm install @mastra/memory
+npm install @mastra/memory @mastra/libsql
 ```
 
 The `@mastra/memory` package provides all the functionality you need to add memory capabilities to your Mastra agents. It includes support for conversation history, semantic recall, and working memory.
 
-This package is separate from the core Mastra package, allowing you to only include it when you need memory capabilities in your project. This modular approach helps keep your project dependencies lean when you don't need all features.
+The `@mastra/libsql` package is **one of many** storage adapters that complement the memory package. The purpose of a storage adapter is to provide a way to persist the memory data to a database or other storage system. The `@mastra/libsql` package provides a storage adapter for LibSQL, which is a fast, open-source fork of SQLite.
+
+These packages are separate from the core Mastra package, allowing you to only include them when you need memory capabilities in your project. This modular approach helps keep your project dependencies lean when you don't need all features.

package/.docs/raw/course/03-agent-memory/04-creating-basic-memory-agent.md

@@ -13,7 +13,7 @@ import { openai } from "@ai-sdk/openai";
 // Create a basic memory instance
 const memory = new Memory({
   storage: new LibSQLStore({
-    url: "../../memory.db",
+    url: "file:../../memory.db", // relative path from the `.mastra/output` directory
   }),
 });
 

package/.docs/raw/course/03-agent-memory/08-configuring-conversation-history.md

@@ -1,6 +1,6 @@
 # Configuring Conversation History
 
-By default, the `Memory` instance includes the last 40 messages from the current memory thread in each new request. You can customize this by configuring the `lastMessages` option:
+By default, the `Memory` instance includes the last 10 messages from the current memory thread in each new request. You can customize this by configuring the `lastMessages` option:
 
 ```typescript
 import { Agent } from "@mastra/core/agent";
@@ -11,10 +11,10 @@ import { LibSQLStore } from "@mastra/libsql";
 // Create a memory instance with custom conversation history settings
 const memory = new Memory({
   storage: new LibSQLStore({
-    url: "../../memory.db",
+    url: "file:../../memory.db",
   }),
   options: {
-    lastMessages: 20, // Include the last 20 messages in the context
+    lastMessages: 20, // Include the last 20 messages in the context instead of the default 10
   },
 });
 

package/.docs/raw/course/03-agent-memory/13-vector-store-configuration.md

@@ -0,0 +1,27 @@
+# Vector Store Configuration
+
+In addition to the memory storage adapters, Mastra also provides vector store adapters useful for storing and retrieving vector embeddings. One of these is the `LibSQLVector` adapter, which provides a simple interface for storing and retrieving vector embeddings in a LibSQL vector database.
+
+```typescript
+import { Memory } from "@mastra/memory";
+import { LibSQLStore, LibSQLVector } from "@mastra/libsql";
+
+const memory = new Memory({
+  storage: new LibSQLStore({
+    url: "file:../../memory.db", // relative path from the `.mastra/output` directory
+  }),
+  vector: new LibSQLVector({
+    connectionUrl: "file:../../vector.db", // relative path from the `.mastra/output` directory
+  }),
+});
+```
+
+Mastra supports several vector store options, including:
+
+- LibSQL
+- Chroma
+- Pinecone
+- Qdrant
+- Postgres (with pgvector)
+
+The vector store is responsible for storing and retrieving the vector embeddings used for semantic search.

package/.docs/raw/course/03-agent-memory/{13-what-is-semantic-recall.md → 14-what-is-semantic-recall.md}

@@ -1,6 +1,6 @@
 # Implementing Semantic Recall
 
-In this step, we'll explore semantic recall, a feature that helps agents maintain context across longer interactions when messages are no longer within recent conversation history.
+One of the quickest ways to take advantage of the configured vector store is to use semantic recall.
 
 ## What is Semantic Recall?
 

package/.docs/raw/course/03-agent-memory/16-configuring-semantic-recall.md

@@ -0,0 +1,41 @@
+# Configuring Semantic Recall
+
+Let's update our agent with custom semantic recall settings:
+
+```typescript
+import { Agent } from "@mastra/core/agent";
+import { Memory } from "@mastra/memory";
+import { openai } from "@ai-sdk/openai";
+import { LibSQLStore, LibSQLVector } from "@mastra/libsql";
+
+// Create a memory instance with semantic recall configuration
+const memory = new Memory({
+  storage: new LibSQLStore({
+    url: "file:../../memory.db", // relative path from the `.mastra/output` directory
+  }), // Storage for message history
+  vector: new LibSQLVector({
+    connectionUrl: "file:../../vector.db", // relative path from the `.mastra/output` directory
+  }), // Vector database for semantic search
+  embedder: openai.embedding("text-embedding-3-small"), // Embedder for message embeddings
+  options: {
+    lastMessages: 20, // Include the last 20 messages in the context
+    semanticRecall: true, // Enable semantic recall with default settings
+  },
+});
+
+// Create an agent with the configured memory
+export const memoryAgent = new Agent({
+  name: "MemoryAgent",
+  instructions: `
+    You are a helpful assistant with advanced memory capabilities.
+    You can remember previous conversations and user preferences.
+    When a user shares information about themselves, acknowledge it and remember it for future reference.
+    If asked about something mentioned earlier in the conversation, recall it accurately.
+    You can also recall relevant information from older conversations when appropriate.
+  `,
+  model: openai("gpt-4o"),
+  memory: memory,
+});
+```
+
+For semantic recall to work, you need to have a **vector store** configured. You also need to have an **embedder** configured. You may use any `@ai-sdk`-compatible embedding model for this. In this example, we're using OpenAI's `text-embedding-3-small` model.

package/.docs/raw/course/03-agent-memory/18-advanced-configuration-semantic-recall.md

@@ -0,0 +1,28 @@
+# Advanced Configuration of Semantic Recall
+
+We can configure semantic recall in more detail by setting options for the `semanticRecall` option:
+
+```typescript
+const memory = new Memory({
+  storage: new LibSQLStore({
+    url: "file:../../memory.db", // relative path from the `.mastra/output` directory
+  }),
+  vector: new LibSQLVector({
+    connectionUrl: "file:../../vector.db", // relative path from the `.mastra/output` directory
+  }),
+  embedder: openai.embedding("text-embedding-3-small"),
+  options: {
+    semanticRecall: {
+      topK: 3,
+      messageRange: {
+        before: 2,
+        after: 1,
+      },
+    },
+  },
+});
+```
+
+The `topK` parameter controls how many semantically similar messages are retrieved. A higher value will retrieve more messages, which can be helpful for complex topics but may also include less relevant information. The default value is `2`.
+
+The `messageRange` parameter controls how much context is included with each match. This is important because the matching message alone might not provide enough context to understand the conversation. Including messages before and after the match helps the agent understand the context of the matched message.

package/.docs/raw/course/03-agent-memory/21-configuring-working-memory.md

@@ -6,11 +6,18 @@ Let's update our agent with working memory capabilities:
 import { Agent } from "@mastra/core/agent";
 import { Memory } from "@mastra/memory";
 import { openai } from "@ai-sdk/openai";
+import { LibSQLStore, LibSQLVector } from "@mastra/libsql";
 
 // Create a memory instance with working memory configuration
 const memory = new Memory({
+  storage: new LibSQLStore({
+    url: "file:../../memory.db", // relative path from the `.mastra/output` directory
+  }), // Storage for message history
+  vector: new LibSQLVector({
+    connectionUrl: "file:../../vector.db", // relative path from the `.mastra/output` directory
+  }), // Vector database for semantic search
+  embedder: openai.embedding("text-embedding-3-small"), // Embedder for message embeddings
   options: {
-    lastMessages: 20,
     semanticRecall: {
       topK: 3,
       messageRange: {
@@ -20,7 +27,6 @@ const memory = new Memory({
     },
     workingMemory: {
       enabled: true,
-      use: "tool-call", // Recommended setting
     },
   },
 });
@@ -52,12 +58,6 @@ export const memoryAgent = new Agent({
 The `workingMemory` configuration has several important options:
 
 - `enabled`: Whether working memory is enabled
-- `use`: How the agent interacts with working memory (recommended setting is "tool-call")
-
-The `use` option can be set to:
-
-- `"tool-call"`: The agent updates working memory via tool calls (recommended)
-- `"direct"`: The agent directly edits the working memory text
-- `"read-only"`: The agent can read but not update working memory
+- `template`: A template for the working memory content
 
 The instructions for the agent are also important. They guide the agent on what information to store in working memory and how to use that information when responding to the user.