Function addFetchInfo

addFetchInfo<T = Record<string, any>>(params: IFetchInfoToolParams<T>): string

Creates and registers a fetch info tool for AI to retrieve data (READ pattern). This implements the READ side of the command pattern - AI calls tool to get information without modifying state.

Flow:

AI calls tool with parameters
validateParams runs (if provided) - validates parameters structure. Returns true if valid, false if invalid
If validation fails (returns false), tool execution is blocked
If validation passes, fetchContent executes - retrieves data
AI receives fetched content as tool output
If content is empty, emptyContent handler is called

Type Parameters

T = Record<string, any>
The type of parameters expected by the fetch operation

Parameters

params: IFetchInfoToolParams<T>
Configuration object for the fetch tool

Returns string

The registered agent tool schema

Function

addFetchInfo

Example

// Fetch user data with parameter validation and error handling
addFetchInfo({
  toolName: "fetch_user_data",
  function: {
    name: "fetch_user_data",
    description: "Fetch user data by user ID",
    parameters: {
      type: "object",
      properties: {
        userId: { type: "string", description: "User ID to fetch" },
      },
      required: ["userId"],
    },
  },
  fallback: (error, clientId, agentName) => {
    logger.error("Failed to fetch user data", { error, clientId, agentName });
  },
  validateParams: async ({ params, clientId, agentName, toolCalls }) => {
    // Returns true if valid, false if invalid
    return !!params.userId;
  },
  fetchContent: async (params, clientId) => {
    const userData = await getUserData(params.userId);
    return JSON.stringify(userData);
  },
  emptyContent: (content) => content || "User not found",
});