*/ toolChoice?: ToolChoice; /** * Maximum number of steps to execute */ maxSteps?: number; /** * System message to include in the prompt * Can be used with `prompt` or `messages` */ system?: string; /** * A simple text prompt * You can either use `prompt` or `messages` but not both */ prompt?: string; /** * A list of messages * You can either use `prompt` or `messages` but not both */ messages?: Array | Array>; /** * Provider-specific options */ providerOptions?: ProviderMetadata; /** * Callback function that is called when a step finishes */ onStepFinish?: GenerateTextOnStepFinishCallback; /** * Function to filter which MCP tools should be available * @param tool The tool to evaluate * @returns Boolean indicating whether to include the tool */ filterMCPTools?: (tool: TOOLS) => boolean; } /** * Tool definition returned by MCP servers * Represents a callable function that the agent can use */ interface MCPTool { /** * Name of the tool */ name: string; /** * Description of what the tool does */ description: string; /** * Parameter schema for the tool */ parameters: { /** * JSON Schema type */ type: string; /** * Properties of the tool parameters */ properties: Record; }; } /** * Type definition for a collection of tools * Maps tool names to their definitions */ type ToolSet = Record; /** * Generic type for tool configurations */ type TOOLS = Record; /** * AI response type to handle OpenAI API responses */ type MCPResponse = GenerateTextResult; ``` ## API ### `AIAgent` The main class that simplifies agent configuration and offers enhanced features. #### Constructor ```typescript constructor({ name, description, systemPrompt, model, verbose, toolsConfigs }: { name: string; description: string; systemPrompt?: string; model?: LanguageModel; verbose?: boolean; toolsConfigs?: WorkflowConfig[]; }) ``` #### Methods ##### `generateResponse(args: GenerateTextArgs): Promise>` Generates a response using the AI model and the tools from the MCP servers and custom tools. Arguments: - `prompt` - The user's message to respond to - `model` - Optional: Override the default model - `system` - Optional: Override the default system prompt - `maxSteps` - Maximum number of steps for the AI to take (default: 20) - `tools` - Additional tools to make available to the model - `filterMCPTools` - Optional function to filter which MCP tools to use - `onStepFinish` - Optional callback function that is called after each step - `messages` - Alternative to prompt, allows for rich content messages ##### `generateObject(args: GenerateObjectArgs): Promise>` Generates a structured object response using a schema. Arguments: - `prompt` - The user's message - `model` - Optional: Override the default model - `schema` - Zod schema or JSON schema for the expected response object - `schemaName` - Optional name for the schema - `schemaDescription` - Optional description for the schema ##### `close(): Promise` Closes all server connections. ##### `getInfo(): { name: string; description: string; tools: string[]; agents?: string[]; model?: LanguageModel; }` Returns information about the agent, including its name, description, available tools, agents, and default model. ## License MIT25:T5ec,## what is MCP AI Agent? MCP AI Agent is a TypeScript library that enables AI agents to leverage Model Context Protocol (MCP) servers for enhanced capabilities, integrating seamlessly with AI SDKs for AI-powered applications. ## how to use MCP AI Agent? To use MCP AI Agent, install it via npm with `npm install mcp-ai-agent`, then import and initialize the agent with preconfigured servers or custom configurations to generate responses using AI models. ## key features of MCP AI Agent? - Connect t

MCP AI Agent

Comments