### Required Setup for Quick Start

Source: https://smithery.ai/docs/use/triggers

To get started, you need a Smithery namespace, an API key, an HTTPS endpoint for webhooks, and an active MCP session.

```plaintext
\"You’ll need:\"
```

```plaintext
\"A Smithery namespace (e.g. \", \"_jsx(_components.code, {\n children: \"my-app\"\n }), \" ) — created on the \", \"_jsx(_components.a, {\n href: \"https://smithery.ai\",\n children: \"dashboard\"\n }), \" or via \", \"_jsx(_components.a, {\n href: \"/use/connect\",\n children: \"Connect\"\n }), \".\"
```

```plaintext
\"A \", \"_jsx(_components.code, {\n children: \"SMITHERY_API_KEY\"\n }), \" from your namespace settings.\"
```

```plaintext
An HTTPS endpoint that can receive webhooks.
```

```plaintext
\"An active MCP session to the connection (via the \", \"_jsx(_components.a, {\n href: \"/use/connect\",\n children: \"SDK\"\n }), \" or CLI).\"
```

--------------------------------

### HTTP Transport Setup Example

Source: https://smithery.ai/docs/use/deep-linking

This example demonstrates setting up a client connection using an HTTP transport. It's suitable for scenarios where HTTP is the preferred communication protocol.

```typescript
// Example with HTTP transport
async function setupHttpMCP(url: string) {
 const config = handleDeepLink(url)
}
```

--------------------------------

### Quick Start: Required Setup

Source: https://smithery.ai/docs/use/triggers

Lists the prerequisites for using Smithery triggers, including namespace creation, API key, and an active MCP session.

```plaintext
You’ll need:
A Smithery namespace (e.g. my-app) — created on the dashboard or via Connect.
A SMITHERY_API_KEY from your namespace settings.
An HTTPS endpoint that can receive webhooks.
An active MCP session to the connection (via the SDK or CLI).
```

--------------------------------

### Example: Stdio MCP Configuration

Source: https://smithery.ai/docs/use/deep-linking

Configuration for connecting to an MCP server using stdio. Requires a command and arguments to start the server executable.

```typescript
const config: MCPConfig = {
  type: "stdio",
  command: "npx",
  args: ["-y", "smithery@latest", "run", "@wonderwhy-er/desktop-commander"]
};
```

--------------------------------

### Setup Stdio Transport for MCP

Source: https://smithery.ai/docs/use/deep-linking

This example shows how to set up a Multi-Channel Protocol (MCP) using the stdio transport. It's designed for scenarios where communication happens over standard input/output streams.

```typescript
// Example with stdio transport
async function setupStdioMCP(url: string
```

--------------------------------

### JavaScript Example for Server Configuration

Source: https://smithery.ai/docs/use/connect

This example demonstrates how to configure a server in JavaScript, passing an API key as a header and a project ID as a query parameter.

```javascript
import { Github } from "@smithery-ai/github";

const github = new Github({
  apiKey: "your-api-key",
  projectId: "your-project-id",
});

const result = await github.searchCode({ query: "example" });
```

--------------------------------

### CLI Connection Example

Source: https://smithery.ai/docs/use/connect

Example of how to initiate a connection using the CLI. This snippet shows a basic command to connect.

```shellscript
smithery connect --url https://github.run.tools
```

--------------------------------

### CLI Example for Server Configuration

Source: https://smithery.ai/docs/use/connect

This example shows how to configure a server using the CLI, passing an API key as a header and a project ID as a query parameter.

```shellscript
export SMITHERY_API_KEY="your-api-key"
mcp \\
  --config "{\"projectId\": \"your-project-id\"}" \\
  github \\
  "search-code" \\
  --query "example"
```

--------------------------------

### Setup Client with Stdio Transport

Source: https://smithery.ai/docs/llms-full.txt

Example implementation for setting up a Smithery AI client using stdio transport based on a deep link configuration. Requires the 'StdioClientTransport' and 'Client' classes.

```typescript
// Example with stdio transport
async function setupStdioMCP(url: string) {
  const config = handleDeepLink(url)
  const transport = new StdioClientTransport({
    command: config.command,
    args: config.args
  })
  
  const client = new Client({ name: "Test client" })
  await client.connect(transport)
  return client
}
```

--------------------------------

### Handling Authorization with Setup URL

Source: https://smithery.ai/docs/use/connect

When initiating an OAuth flow, use the provided setup URL to redirect users. This ensures a stable connection ID for subsequent retries.

```javascript
When you need to send a user through OAuth, prefer creating or updating the connection first and redirecting to the hosted ", _jsx(_components.code, {
  children: "setupUrl"
}), ". That gives you a stable ", _jsx(_components.code, {
  children: "connectionId"
}), " to retry with after the user returns to your app."
```

--------------------------------

### CLI Example: Aggregate Tools Across Connections

Source: https://smithery.ai/docs/use/connect

This example shows how to aggregate tools from multiple connections using the CLI. It's useful for managing and interacting with tools from various service endpoints.

```shellscript
smithery mcp --connections "$(smithery conn)"
```

--------------------------------

### Stdio-based Configuration Example

Source: https://smithery.ai/docs/use/deep-linking

An example JSON configuration for a stdio-based server connection. Specifies the type, command, and arguments for the server executable.

```json
{
 "type": "stdio",
 "command": "npx",
 "args": [
  "-y",
  "smithery@latest",
  "run",
  "@wonderwhy-er/desktop-commander"
 ]
}
```

--------------------------------

### Header Transformation Example

Source: https://smithery.ai/docs/build/session-config

This example demonstrates how to map a client-provided header 'x-from' to a server-expected header 'Authorization'. It also shows the mapping for 'x-to' to 'Authorization'.

```javascript
_.jsxs(_.components.span, {
 className: "line",
 children: [
  _jsx(_.components.span, {
   style: {
    color: "#0A3069",
    "--shiki-dark": "#CE9178"
   },
   children: " \"x-from\""
  }),
  _jsx(_.components.span, {
   style: {
    color: "#1F2328",
    "--shiki-dark": "#9CDCFE"
   },
   children: ":"
  }),
  _jsx(_.components.span, {
   style: {
    color: "#1F2328",
    "--shiki-dark": "#D4D4D4"
   },
   children: " { "
  }),
  _jsx(_.components.span, {
   style: {
    color: "#1F2328",
    "--shiki-dark": "#9CDCFE"
   },
   children: "header:"
  }),
  _jsx(_.components.span, {
   style: {
    color: "#0A3069",
    "--shiki-dark": "#CE9178"
   },
   children: " \"posthog-api-key\""
  }),
  _jsx(_.components.span, {
   style: {
    color: "#1F2328",
    "--shiki-dark": "#D4D4D4"
   },
   children: " }, "
  }),
  _jsx(_.components.span, {
   style: {
    color: "#6E7781",
    "--shiki-dark": "#6A9955"
   },
   children: "// Client provides this header"
  })
 ]
}), "\n", _jsxs(_.components.span, {
 className: "line",
 children: [
  _jsx(_.components.span, {
   style: {
    color: "#0A3069",
    "--shiki-dark": "#CE9178"
   },
   children: " \"x-to\""
  }),
  _jsx(_.components.span, {
   style: {
    color: "#1F2328",
    "--shiki-dark": "#9CDCFE"
   },
   children: ":"
  }),
  _jsx(_.components.span, {
   style: {
    color: "#1F2328",
    "--shiki-dark": "#D4D4D4"
   },
   children: " { "
  }),
  _jsx(_.components.span, {
   style: {
    color: "#1F2328",
    "--shiki-dark": "#9CDCFE"
   },
   children: "header:"
  }),
  _jsx(_.components.span, {
   style: {
    color: "#0A3069",
    "--shiki-dark": "#CE9178"
   },
   children: " \"Authorization\""
  }),
  _jsx(_.components.span, {
   style: {
    color: "#1F2328",
    "--shiki-dark": "#D4D4D4"
   },
   children: " } "
  }),
  _jsx(_.components.span, {
   style: {
    color: "#6E7781",
    "--shiki-dark": "#6A9955"
   },
   children: "// PostHog expects Authorization"
  })
 ]
}), "\n", _jsx(_.components.span, {
 className: "line",
 children: _jsx(_.components.span, {
  style: {
   color: "#1F2328",
   "--shiki-dark": "#D4D4D4"
  },
  children: " })"
 })
]), "\n", _jsxs(_.components.span, {
 className: "line",
 children: [
  _jsx(_.components.span, {
   style: {
    color: "#1F2328",
    "--shiki-dark": "#D4D4D4"
   },
   children: " ."
  }),
  _jsx(_.components.span, {
   style: {
    color: "#8250DF",
    "--shiki-dark": "#DCDCAA"
   },
   children: "describe"
  }),
  _jsx(_.components.span, {
   style: {
    color: "#1F2328",
    "--shiki-dark": "#D4D4D4"
   },
   children: "("
  }),
  _jsx(_.components.span, {
   style: {
    color: "#0A3069",
    "--shiki-dark": "#CE9178"
   },
   children: " \"Your PostHog API key\""
  }),
  _jsx(_.components.span, {
   style: {
    color: "#1F2328",
    "--shiki-dark": "#D4D4D4"
   },
   children: "),"
  })
 ]
}), "\n", _jsx(_.components.span, {
 className: "line",
 children: _jsx(_.components.span, {
  style: {
   color: "#1F2328",
   "--shiki-dark": "#D4D4D4"
  },
  children: "});"
 })
])

```

--------------------------------

### Read-Only Dashboard Token CLI Example

Source: https://smithery.ai/docs/use/token-scoping

This example demonstrates how to create a read-only token for dashboard access using the CLI.

```shellscript
smithery token create --resource connections --operations read --ttl 1h
```

--------------------------------

### Install Typed SDK for Exa

Source: https://smithery.ai/docs/llms-full.txt

Installs the typed TypeScript SDK for the Exa MCP server from the Smithery package registry.

```bash
npm install https://pkg.smithery.ai/exa
```

--------------------------------

### Install Smithery CLI

Source: https://smithery.ai/docs/llms-full.txt

Install the Smithery CLI globally using npm. Requires Node.js 20 or later.

```bash
npm install -g smithery@latest
```

--------------------------------

### Metadata AND Logic Example

Source: https://smithery.ai/docs/use/token-scoping

Demonstrates how metadata within a single object is ANDed. Use a list for OR logic.

```javascript
{
  owner: 'alice',
  env: 'prod'
}
```

```javascript
[{ owner: 'alice' }, { env: 'prod' }]
```

--------------------------------

### Smithery CLI Examples

Source: https://smithery.ai/docs/concepts/cli

Common Smithery CLI commands for viewing help, authenticating, and searching for MCP servers.

```shellscript
# Show the help menu
smithery --help
```

```shellscript
# Authenticate with Smithery
smithery auth login
```

```shellscript
# Search for MCP servers
smithery
```

--------------------------------

### Search for skills

Source: https://smithery.ai/docs/concepts/cli

Use this command to search for available skills. The example searches for skills related to 'code review'.

```bash
smithery skill search "code review"
```

--------------------------------

### Example Connection Configuration

Source: https://smithery.ai/docs/use/connect

This snippet shows how to configure a connection with a specific namespace and connection ID. Use this for establishing a connection to a service.

```javascript
const exa = new Exa(
  {
    apiKey: "YOUR_API_KEY",
    // You can also pass in a connection ID
    connectionId: "exa",
    // Or a namespace
    namespace: "my-app",
  }
);

```

--------------------------------

### Adding a Server with Configuration via CLI

Source: https://smithery.ai/docs/use/connect

Shows how to add a server using the CLI, specifying configuration parameters like an API key as a header and a project ID as a query parameter.

```shellscript
# Add a server with config (API key as header, project ID as query param)
smithery server add github \"https://api.github.com\" \"--header\" \"Authorization: token $GITHUB_API_TOKEN\" \"--query\" \"client_id=$GITHUB_CLIENT_ID\"
```

--------------------------------

### Setup Client with HTTP Transport

Source: https://smithery.ai/docs/llms-full.txt

Example implementation for setting up a Smithery AI client using HTTP transport based on a deep link configuration. Requires the 'StreamableHTTPClientTransport' and 'Client' classes.

```typescript
// Example with HTTP transport
async function setupHttpMCP(url: string) {
  const config = handleDeepLink(url)
  const transport = new StreamableHTTPClientTransport(config.url)
  
  const client = new Client({ name: "Test client" })
  await client.connect(transport)
  return client
}
```

--------------------------------

### CLI Connection Example

Source: https://smithery.ai/docs/use/connect

Use this command to establish a connection to a service via the CLI. Ensure you have the necessary authentication tokens.

```shellscript
# Connect to a
```

--------------------------------

### Get a namespace

Source: https://smithery.ai/docs/api-reference/namespaces

Get a single namespace by its name.

```APIDOC
## GET /namespaces/{name}

### Description
Get a single namespace by its name.

### Method
GET

### Endpoint
/namespaces/{name}

### Parameters
#### Path Parameters
- **name** (string) - Required - The name of the namespace to retrieve.

### Response
#### Success Response (200)
- **name** (string) - The name of the namespace.
- **owner** (string) - The owner of the namespace.
- **createdAt** (string) - The timestamp when the namespace was created.
```

--------------------------------

### CLI Quick Start for Smithery

Source: https://smithery.ai/docs/use/connect

Basic CLI commands to log in, add a server, list tools, and call a tool.

```bash
smithery auth login

smithery mcp add exa --id exa

smithery tool list exa

smithery tool call exa search '{"query": "latest news about MCP"}'
```

--------------------------------

### Get a skill

Source: https://smithery.ai/docs/api-reference/skills

Get a single skill by its namespace and slug.

```APIDOC
## GET /skills/{namespace}/{slug}

### Description
Get a single skill by its namespace and slug.

### Method
GET

### Endpoint
/skills/{namespace}/{slug}

### Parameters
#### Path Parameters
- **namespace** (string) - Required - The namespace of the skill.
- **slug** (string) - Required - The slug of the skill.

### Response
#### Success Response (200)
- **title** (string) - The title of the skill.
- **description** (string) - A brief description of the skill.
- **namespace** (string) - The namespace the skill belongs to.
- **slug** (string) - The unique slug for the skill.
- **gitUrl** (string) - The GitHub repository URL for the skill.

#### Response Example
{
  "title": "Summarize Text",
  "description": "Summarizes a given piece of text.",
  "namespace": "example",
  "slug": "summarize-text",
  "gitUrl": "https://github.com/example/summarize-text.git"
}
```

--------------------------------

### Create a Token for Reading Servers

Source: https://smithery.ai/docs/use/token-scoping

This example shows how to create a token that allows read access to servers within the 'my-app' namespace. The token has a TTL of 1 hour.

```bash
curl -X POST https://api.smithery.ai/v1/tokens \ 
 -H "Authorization: Bearer $SMITHERY_API_KEY" \ 
 -H "Content-Type: application/json" \ 
 -d '{ 
 "namespaces": "my-app", 
 "resources": "servers", 
 "operations": "read", 
 "ttl": "1h" 
 }'
```

--------------------------------

### GET Get a server

Source: https://smithery.ai/docs/api-reference/servers

Retrieves details for a specific server.

```APIDOC
## GET /servers/{server_id}

### Description
Retrieves details for a specific server.

### Method
GET

### Endpoint
/servers/{server_id}

### Parameters
#### Path Parameters
- **server_id** (string) - Required - The unique identifier of the server.
```

--------------------------------

### Get Trigger

Source: https://smithery.ai/docs/concepts/registry_search_servers

Gets the schema for a single trigger type.

```APIDOC
## GET /connect/{namespace}/{connectionId}/.triggers/{triggerName}

### Description
Gets the schema for a single trigger type.

### Method
GET

### Endpoint
/connect/{namespace}/{connectionId}/.triggers/{triggerName}

### Parameters
#### Path Parameters
- **namespace** (string) - Required - The namespace of the connection.
- **connectionId** (string) - Required - The ID of the connection.
- **triggerName** (string) - Required - The name of the trigger.

### Request Body
(No request body details provided in the source)

### Response
#### Success Response (200)
(No response details provided in the source)

#### Response Example
(No response example provided in the source)
```

--------------------------------

### Create a Multi-Resource Token

Source: https://smithery.ai/docs/use/token-scoping

This example shows how to create a token that grants access to multiple resources ('connections' and 'databases') and multiple operations ('execute' and 'read').

```shell
smithery token create \
  -H "Content-Type: application/json" \
  -d '{
    "policy": [
      {
        "namespaces": "my-app",
        "resources": ["connections", "databases"],
        "operations": ["execute", "read"],
        "metadata": { "userId": "user-123" },
        "ttl": "30m"
      }
    ]
  }'
```

--------------------------------

### Server Configuration Example

Source: https://smithery.ai/docs/build/session-config

This snippet shows how to define server configuration, including its name and version. It's used to set up the basic parameters for a session.

```javascript
const config = {
  server: {
    name: "My Server",
    version: "1.0.0",
  },
};

```

--------------------------------

### GET Get a release

Source: https://smithery.ai/docs/api-reference/servers

Retrieves details for a specific release of a server.

```APIDOC
## GET /servers/{server_id}/releases/{release_id}

### Description
Retrieves details for a specific release of a server.

### Method
GET

### Endpoint
/servers/{server_id}/releases/{release_id}

### Parameters
#### Path Parameters
- **server_id** (string) - Required - The unique identifier of the server.
- **release_id** (string) - Required - The unique identifier of the release.
```

--------------------------------

### Get Tool

Source: https://smithery.ai/docs/concepts/registry_search_servers

Gets one tool or lists tools under a slash-separated category.

```APIDOC
## GET /connect/{namespace}/{connectionId}/.tools/{toolName}

### Description
Gets one tool or lists tools under a slash-separated category.

### Method
GET

### Endpoint
/connect/{namespace}/{connectionId}/.tools/{toolName}

### Parameters
#### Path Parameters
- **namespace** (string) - Required - The namespace of the connection.
- **connectionId** (string) - Required - The ID of the connection.
- **toolName** (string) - Required - The name of the tool or category.

### Request Body
(No request body details provided in the source)

### Response
#### Success Response (200)
(No response details provided in the source)

#### Response Example
(No response example provided in the source)
```

--------------------------------

### Advanced Smithery CLI Examples

Source: https://smithery.ai/docs/llms-full.txt

Demonstrates advanced usage of the Smithery CLI, including adding servers with configurations, JSON output, and publishing.

```bash
# Add an MCP server to Claude Desktop
smithery mcp add mcp-obsidian --client claude
```

```bash
# Add with pre-configured data (skips prompts)
smithery mcp add mcp-obsidian --client claude --config '{"vaultPath":"path/to/vault"}'
```

```bash
# Remove a server from a client
smithery mcp remove mcp-obsidian --client claude
```

```bash
# Search for MCP servers with JSON output
smithery --json mcp search "database"
```

```bash
# List tools from a specific connection
smithery tool list my-github
```

```bash
# Find tools by intent
smithery tool find "create issue"
```

```bash
# Call a tool with JSON arguments
smithery tool call my-github create_issue '{"title":"Bug fix","body":"..."}'
```

```bash
# Login and check auth
smithery auth login
smithery auth whoami
```

```bash
# Publish your MCP server
smithery mcp publish "https://my-server.com" -n myorg/my-server
```

```bash
# Publish an MCPB bundle
smithery mcp publish ./server.mcpb -n myorg/my-server
```

```bash
# Show help
smithery --help
```

--------------------------------

### Example Trigger Payload

Source: https://smithery.ai/docs/use/triggers

This is an example of a trigger payload that might be received. It contains an array of events.

```json
{
"events": [
 {
 "id": "trigger-id-1",
 "name": "example.trigger.name",
 "description": "This is an example trigger.",
 "inputSchema": {
 "type": "object",
 "properties": {
 "message": {
 "type": "string"
 },
 "count": {
 "type": "integer"
 }
 },
 "required": [
 "message"
 ]
 },
 "outputSchema": {
 "type": "object",
 "properties": {
 "status": {
 "type": "string"
 },
 "processed_count": {
 "type": "integer"
 }
 },
 "required": [
 "status"
 ]
 },
 "tags": [
 "example",
 "test"
 ]
 },
 {
 "id": "trigger-id-2",
 "name": "another.trigger.name",
 "description": "Another example trigger.",
 "inputSchema": {
 "type": "object",
 "properties": {
 "data": {
 "type": "array",
 "items": {
 "type": "string"
 }
 }
 },
 "required": [
 "data"
 ]
 },
 "outputSchema": {
 "type": "object",
 "properties": {
 "success": {
 "type": "boolean"
 }
 },
 "required": [
 "success"
 ]
 },
 "tags": [
 "example"
 ]
 }
]
}
```

--------------------------------

### Get Connection

Source: https://smithery.ai/docs/concepts/registry_search_servers

Gets details for a specific connection. Requires service token with connections:read scope.

```APIDOC
## GET /connect/{namespace}/{connectionId}

### Description
Gets details for a specific connection. Requires service token with connections:read scope.

### Method
GET

### Endpoint
/connect/{namespace}/{connectionId}

### Parameters
#### Path Parameters
- **namespace** (string) - Required - The namespace of the connection.
- **connectionId** (string) - Required - The ID of the connection.

### Request Body
(No request body details provided in the source)

### Response
#### Success Response (200)
(No response details provided in the source)

#### Response Example
(No response example provided in the source)
```

--------------------------------

### Call Tool Example

Source: https://smithery.ai/docs/use/connect

Demonstrates how to call a tool named 'search_repositories' with specific arguments using the MCP client. This snippet shows the structure for making tool calls.

```javascript
_jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#9CDCFE"
 },
 children: " mcpClient"
 }), _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#D4D4D4"
 },
 children: "."
 }), _jsx(_components.span, {
 style: {
 color: "#8250DF",
 "--shiki-dark": "#DCDCAA"
 },
 children: "callTool"
 }), _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#D4D4D4"
 },
 children: "({"
 }) , "\n", _jsxs(_components.span, {
 className: "line",
 children: [_jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#9CDCFE"
 },
 children: " name:"
 }), _jsx(_components.span, {
 style: {
 color: "#0A3069",
 "--shiki-dark": "#CE9178"
 },
 children: " 'search_repositories'"
 }), _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#D4D4D4"
 },
 children: ","
 }) ]
 }), "\n", _jsxs(_components.span, {
 className: "line",
 children: [_jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#9CDCFE"
 },
 children: " arguments:"
 }), _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#D4D4D4"
 },
 children: " { "
 }), _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#9CDCFE"
 },
 children: "query:"
 }), _jsx(_components.span, {
 style: {
 color: "#0A3069",
 "--shiki-dark": "#CE9178"
 },
 children: " 'mcp'"
 }), _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#D4D4D4"
 },
 children: " },"
 }) ]
 }), "\n", _jsx(_components.span, {
 className: "line",
 children: _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#D4D4D4"
 },
 children: "})"
 }) )
 })

```

--------------------------------

### CLI Quick Start: Connect to Exa Search

Source: https://smithery.ai/docs/use/connect

This snippet shows the command-line steps to log in to Smithery and connect to the Exa search server using the CLI.

```shellscript
# 1. Log in to Smithery
smithery auth login

# 2. Connect to the Exa search server
smithery mcp add exa --id exa
```

--------------------------------

### Example Deep Link

Source: https://smithery.ai/docs/llms-full.txt

An example illustrating the deep link format with a specific client scheme and handler.

```typescript
// Example:
// cursor://anysphere.cursor-deeplink/mcp/install?<server-display-name>&<base64-encoded-json-config>
```

--------------------------------

### Example Tool and Prompt Configuration

Source: https://smithery.ai/docs/use/connect

This snippet shows how to configure tools and prompts for use with Smithery AI. It demonstrates the structure for defining available tools and the user's message.

```javascript
_jsxs(_components.span, {
 className: "line",
 children: [_jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#9CDCFE"
 },
 children: " tools:"
 }), _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#9CDCFE"
 },
 children: " allTools"
 }), _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#D4D4D4"
 },
 children: ","
 })]
 }), "\n", _jsxs(_components.span, {
 className: "line",
 children: [_jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#9CDCFE"
 },
 children: " prompt:"
 }), _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#9CDCFE"
 },
 children: " userMessage"
 }), _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#D4D4D4"
 },
 children: ","
 })]
 }), "\n", _jsx(_components.span, {
 className: "line",
 children: _jsx(_components.span, {
 style: {
 color: "#1F2328",
 "--shiki-dark": "#D4D4D4"
 },
 children: "});"
 })
 }), "\n"]
 })
)
```

--------------------------------

### Install AI SDK packages

Source: https://smithery.ai/docs/llms-full.txt

Installs necessary npm packages for using the AI SDK with MCP and Anthropic.

```bash
npm install @smithery/api @ai-sdk/mcp ai @ai-sdk/anthropic
```

--------------------------------

### Show Help Menu

Source: https://smithery.ai/docs/concepts/cli

Displays the main help menu for the Smithery CLI.

```bash
smithery --help
```

--------------------------------

### Client Initialization and Connection

Source: https://smithery.ai/docs/use/deep-linking

Demonstrates how to initialize a client and establish a connection using the provided transport. This is a common setup for interacting with services.

```typescript
const client = new Client({
  name: "Test client"
});

await client.connect(transport)

return client
```

--------------------------------

### Example Trigger Payload

Source: https://smithery.ai/docs/use/triggers

This is an example of a trigger event payload that might be received. It contains an 'events' array with details about the trigger.

```json
{
"events": [
 {
 "event": "example.trigger.name",
 "description": "This is an example trigger.",
 "input": {
 "type": "object",
 "properties": {
 "message": {
 "type": "string"
 },
 "count": {
 "type": "integer"
 }
 },
 "required": [
 "message"
 ]
 },
 "output": {
 "type": "object",
 "properties": {
 "status": {
 "type": "string"
 },
 "processed_count": {
 "type": "integer"
 }
 },
 "required": [
 "status"
 ]
 }
 }
]
}
```

--------------------------------

### Http-based Configuration Example

Source: https://smithery.ai/docs/use/deep-linking

An example JSON configuration for an http-based server connection. Specifies the type and the URL of the MCP server.

```json
{
 "type": "http",
 "url": "https://exa.run.tools"
}
```

--------------------------------

### Using MCP Client in JavaScript

Source: https://smithery.ai/docs/use/connect

Example of initializing and using the MCP client in JavaScript. This snippet demonstrates how to set up the client and close the connection.

```javascript
tools, 
 prompt: 'Search for the latest news about MCP.',
});
await mcpClient.close();
```

--------------------------------

### Create a Single-Resource Token

Source: https://smithery.ai/docs/use/token-scoping

This example demonstrates how to create a token with a policy that grants access to a single resource ('connections') and a single operation ('execute').

```shell
smithery token create \
  -H "Content-Type: application/json" \
  -d '{
    "policy": [
      {
        "namespaces": "my-app",
        "resources": "connections",
        "operations": "execute",
        "metadata": { "userId": "user-123" },
        "ttl": "30m"
      }
    ]
  }'
```

--------------------------------

### Install MCP TypeScript SDK packages

Source: https://smithery.ai/docs/llms-full.txt

Installs the Smithery API and the Model Context Protocol (MCP) SDK client packages.

```bash
npm install @smithery/api @modelcontextprotocol/sdk
```

--------------------------------

### Install Vercel AI SDK Dependencies

Source: https://smithery.ai/docs/llms-full.txt

Install the necessary packages for Vercel AI SDK, MCP tools, Anthropic, and Smithery API.

```bash
npm install ai @ai-sdk/mcp @ai-sdk/anthropic @smithery/api
```

--------------------------------

### TypeScript Connection Setup

Source: https://smithery.ai/docs/use/connect

This TypeScript snippet demonstrates how to set up a Browserbase connection. It includes setting the namespace and MCP URL.

```typescript
const conn = await smithery.connections.set(
  'my-browserbase', { 
 namespace: 'my-app',
 mcpUrl: 'https://mcp.browserbase.com/mcp'
```

--------------------------------

### Initialize and Use MCP TypeScript SDK

Source: https://smithery.ai/docs/use/connect

A comprehensive example of initializing the MCP client and using it to generate text. This snippet covers setting up the client and making a request.

```typescript
import { MCPClient } from "@modelcontextprotocol/sdk";

const mcpClient = new MCPClient({
  // ... client configuration options
});

async function main() {
  const response = await mcpClient.generateText({
    model: "anthropic"(
      'claude-sonnet-4-20250514'
    ),
    tools,
    prompt: 'Search for the latest news about MCP.',
  });

  console.log(response);

  await mcpClient.close();
}

main();
```

--------------------------------

### TypeScript Example for Token Scoping

Source: https://smithery.ai/docs/use/token-scoping

A TypeScript example demonstrating the structure for defining token scopes. This code illustrates how to programmatically construct the scope object.

```typescript
const scope = {
  namespaces: "my-app",
  resources: "connections",
  operations: ["read", "execute"],
  metadata: { "userId": "user-123" },
  ttl: "1h"
};
```

--------------------------------

### Initialize Smithery Client

Source: https://smithery.ai/docs/use/connect

Instantiate the Smithery client. This is the first step before establishing any connections.

```javascript
const smithery = new Smithery();
```

--------------------------------

### Creating MCP Clients for Connections (CLI)

Source: https://smithery.ai/docs/use/connect

This shell script example shows how to create MCP clients for each connection using the Smithery CLI. This is useful for managing multiple connections and their associated tools.

```shellscript
smithery mcp --connection my-connection --connection another-connection
```

--------------------------------

### Mint Token with OR Logic for Resources

Source: https://smithery.ai/docs/use/token-scoping

This example demonstrates minting a token that grants read access to 'alice's connections' OR read/write access to 'servers'. Metadata within a single object is AND’d, while a list of objects enables OR logic.

```shell
curl -X POST https://api.smithery.ai/tokens \
  -H "Authorization: Bearer $SMITHERY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "policy": [
      {
        "resources": "connections",
        "operations": "read",
        "metadata": { "owner": "alice" }
      },
      {
        "resources": "servers",
        "operations": ["read", "write"]
      }
    ]
  }'
```

--------------------------------

### Example Deep Link

Source: https://smithery.ai/docs/use/deep-linking

An illustrative example of a deep link using the 'cursor' scheme, demonstrating the expected structure with a display name and encoded configuration.

```plaintext
cursor://anysphere.cursor-deeplink/mcp/install?<server-display-name>&<base64-encoded-json-config>
```

--------------------------------

### Handling Authentication Required

Source: https://smithery.ai/docs/use/connect

When authentication is required, the CLI will output a setup URL. Visit this URL to authorize the connection, then retry the operation.

```text
# If auth is required, the CLI outputs the setup URL:
# → auth_required
# → https://auth.smithery.ai/...
# → connection_id: abc-123-github
# Visit the URL to authorize, then retry
```

--------------------------------

### Create Multi-Resource Token

Source: https://smithery.ai/docs/use/token-scoping

This example shows how to create a single token that grants access to multiple resources with different operations. This simplifies access management by consolidating permissions.

```bash
smithery auth token --policy '[
  {
    "namespaces": "my-app",
    "resources": "connections",
    "operations": ["read", "execute"],
    "metadata": { "userId": "user-123" },
    "ttl": "1h"
  },
  {
    "namespaces": "my-app",
    "resources": "servers",
    "operations": "read",
    "ttl": "1h"
  }
]'
```

--------------------------------

### Get Connection Details with Smithery CLI

Source: https://smithery.ai/docs/concepts/cli

Use the 'smithery mcp get <i>' command to retrieve detailed information about a specific connection. Replace <i> with the connection's ID.

```bash
smithery mcp get <i> # Get connection details
```

--------------------------------

### Set Up a New Connection

Source: https://smithery.ai/docs/use/connect

Establish a connection to a specific Smithery namespace. Ensure the `mcpUrl` includes your `browserbaseProjectId`.

```javascript
const conn = await smithery.connections.set(
  'my-browserbase',
  {
    namespace: 'my-app',
    mcpUrl: 'https://mcp.browserbase.com/mcp?browserbaseProjectId=your-project-id',
  }
);
```

--------------------------------

### Example Token Scope Definition

Source: https://smithery.ai/docs/use/token-scoping

Defines the scope for a token, specifying allowed resources and operations. This example shows a scope for 'servers' resource with 'read' and 'write' operations.

```json
{
  "resources": {
    "servers": {
      "operations": ["read", "write"]
    }
  }
}
```

--------------------------------

### Execute-Only Agent Token CLI Example

Source: https://smithery.ai/docs/use/token-scoping

This example demonstrates how to create an agent token with execute-only permissions using the CLI. Ensure your SMITHERY_API_KEY environment variable is set.

```shellscript
curl -X POST \
  https://api.smithery.ai/tokens \ 
  -H "Authorization: Bearer $SMITHERY_API_KEY" \ 
  -H "Content-Type: application/json" \ 
  -d '{ 
 "policy": [ 
 { 
 "namespaces": "my-app", 
 "resources": "connections", 
 "operations": "execute", 
 "ttl": "1h" 
 } 
 ] 
 }'
```

--------------------------------

### Example Tool Search

Source: https://smithery.ai/docs/use/connect

This snippet demonstrates how to use the search tool with a query. The SDK parses and validates structured data returned by tools.

```javascript
const result = await exa.tools.search({
  query: "latest news about MCP",
});

```

--------------------------------

### Example Token with Scopes

Source: https://smithery.ai/docs/use/token-scoping

This example demonstrates a JSON Web Token (JWT) with defined scopes for resources and operations. It includes metadata like userId and a time-to-live (ttl).

```json
{
  "policy": [
    {
      "resources": "connections",
      "operations": "read",
      "metadata": {"userId": "user-123"},
      "ttl": "20m"
    }
  ]
}
```

--------------------------------

### Create a Token with Specific Scopes

Source: https://smithery.ai/docs/use/token-scoping

This example demonstrates how to create a token with a policy that defines access to specific namespaces, resources, and operations. The policy includes metadata and a time-to-live (TTL).

```shell
curl -X POST https://api.example.com/v1/tokens \
 -H "Content-Type: application/json" \
 -d '{
 "policy": [
  {
  "namespaces": "my-app",
  "resources": "connections",
  "operations": ["read", "execute"],
  "metadata": { "userId": "user-123", "tier": "pro" },
  "ttl": "1h"
 }
 ]
 }'
```

--------------------------------

### Browserbase MCP Connection Example

Source: https://smithery.ai/docs/use/connect

This snippet demonstrates how to establish an MCP connection to Browserbase, specifying the project ID, and providing API key headers.

```shell
smithery mcp add '@browserbasehq/mcp-browserbase?browserbaseProjectId=your-project-id' \"\"
--id my-browserbase
--headers '{"browserbaseApiKey": "your-browserbase-api-key"}'
```

--------------------------------

### Example Trigger Schema

Source: https://smithery.ai/docs/use/triggers

This is an example response showing the schema for an event trigger, including its name, description, delivery options, and input/payload schemas. Use this to understand event structure.

```json
{
  "events": [
    {
      "name": "page.updated",
      "description": "Fires when a page in the watched workspace is updated.",
      "delivery": ["webhook"],
      "inputSchema": {
        "type": "object",
        "properties": {
          "workspace_id": { "type": "string" }
        },
        "required": ["workspace_id"]
      },
      "payloadSchema": {
        "type": "object",
        "properties": {
          "page_id": { "type": "string" },
          "updated_at": { "type": "string", "format": "date-time" }
        }
      }
    }
  ]
}
```

--------------------------------

### Direct Client Connection Example

Source: https://smithery.ai/docs/use/deep-linking

This snippet shows how to establish a direct connection to a client using a transport. It's useful for setting up real-time communication channels.

```typescript
async function setupDirectClient(transport: string) {
  const client = await client.connect(transport)
  return client
}
```

--------------------------------

### Get Full Details for a Specific Tool

Source: https://smithery.ai/docs/concepts/cli

Retrieve comprehensive information about a particular tool using 'smithery tool get <tool_name>'. This command displays all available details for the specified tool.

```shellscript
smithery tool get <connection> # Show full details for one tool
```

--------------------------------

### JavaScript Connection Setup

Source: https://smithery.ai/docs/use/connect

Use this JavaScript snippet to set up a connection to Browserbase. It specifies the connection ID, namespace, and MCP URL.

```javascript
_jsx(_components.span, {
 style: {
 color: "#0A3069",
 "--shiki-dark": "#CE9178"
 },
 children: " add"
 }), _jsx(_components.span, {
 style: {
 color: "#CF222E",
 "--shiki-dark": "#D7BA7D"
 },
 children: " \\\\""
 })] 
 }), "\\n", _jsxs(_components.span, {
 className: "line",
 children: [_jsx(_components.span, {
 style: {
 color: "#0A3069",
 "--shiki-dark": "#CE9178"
 },
 children: " 'https://mcp.browserbase.com/mcp?browserbaseProjectId=your-project-id'"
 }), _jsx(_components.span, {
 style: {
 color: "#CF222E",
 "--shiki-dark": "#D7BA7D"
 },
 children: " \\\\""
 })] 
 }), "\\n", _jsxs(_components.span, {
 className: "line",
 children: [_jsx(_components.span, {
 style: {
 color: "#0550AE",
 "--shiki-dark": "#569CD6"
 },
 children: " --id"
 }), _jsx(_components.span, {
 style: {
 color: "#0A3069",
 "--shiki-dark": "#CE9178"
 },
 children: " my-browserbase"
 }), _jsx(_components.span, {
 style: {
 color: "#CF222E",
 "--shiki-dark": "#D7BA7D"
 },
 children: " \\\\""
 })] 
 }), "\\n", _jsxs(_components.span, {
 className: "line",
 children: [_jsx(_components.span, {
 style: {
 color: "#0550AE",
 "--shiki-dark": "#569CD6"
 },
 children: " --headers"
 }), _jsx(_components.span, {
 style: {
 color: "#0A3069",
 "--shiki-dark": "#CE9178"
 },
 children: " '{\\\"browserbaseApiKey\\\": \\\
```

--------------------------------

### Deep Link Example URL

Source: https://smithery.ai/docs/use/deep-linking

This is an example of a complete deep link URL for the 'mcp/install' path. It shows the expected format with placeholders for server display name and base64-encoded JSON configuration.

```text
// cursor://anysphere.cursor-deeplink/mcp/install?<server-display-name>&<base64-encoded-json-config>
```

--------------------------------

### Initialize MCP Client

Source: https://smithery.ai/docs/use/connect

Instantiate the MCP client with a name and version. This is the first step to interacting with the MCP service.

```javascript
const mcpClient = new Client({
  name: 'my-app',
  version: '1.0.0'
});
```

--------------------------------

### Install Next.js and MCP TypeScript SDK

Source: https://smithery.ai/docs/llms-full.txt

Installs the necessary dependencies for a new Next.js project and the MCP TypeScript SDK. Ensure you are in your project's root directory before running these commands.

```bash
npx create-next-app@latest my-mcp-client
cd my-mcp-client
npm install @modelcontextprotocol/sdk
```

--------------------------------

### Add Server with API Key and Project ID (Shell)

Source: https://smithery.ai/docs/use/connect

Use this command to add a new server connection. Configure API key as a header and project ID as a query parameter.

```shellscript
# Add a server with config (API key as header, project ID as query param)
smithery mcp add \"@browserbasehq/mcp-browserbase?browserbaseProjectId=your-project-id\" \" --id my-browserbase --headers '{\"browserbaseApiKey\": \"your-browserbase-api-key\"}'
```

--------------------------------

### Create MCP client and generate text with AI SDK

Source: https://smithery.ai/docs/llms-full.txt

Demonstrates creating an MCP client and using the AI SDK to generate text, leveraging MCP tools. Ensure you have the necessary SDKs installed.

```typescript
import { createMCPClient } from '@ai-sdk/mcp';
import { generateText } from 'ai';
import { anthropic } from '@ai-sdk/anthropic';
import { createConnection } from '@smithery/api/mcp';

const { transport } = await createConnection({
  mcpUrl: 'https://mcp.exa.ai',
});

const mcpClient = await createMCPClient({ transport });
const tools = await mcpClient.tools();

const { text } = await generateText({
  model: anthropic('claude-sonnet-4-20250514'),
  tools,
  prompt: 'Search for the latest news about MCP.',
});

await mcpClient.close();
```

--------------------------------

### TypeScript Example: Specifying Time-Based Scopes

Source: https://smithery.ai/docs/use/token-scoping

Demonstrates how to define time-based scopes for tokens, allowing for temporary access. This example shows how to specify a duration like '1h', '30m', or '20s', or a specific number of seconds.

```typescript
// ttl: time-to-live
{
  ttl: "1h" | "30m" | "20s" | number // e.g., \"1h\", \"30m\", \"20s\", 3600
}
```