Return finished web results inside Mastra.

Why Mastra + Tabstack

Adding web access to an agent usually means writing your own fetch and parse code. @tabstack/mastra replaces that with a hosted API exposed as typed createTool definitions: define the shape with Zod and get that shape back, with JS-heavy pages rendered server-side and one key for extraction, research, generation, and automation.

Quickstart

Install the adapter and set your key:

npm install @tabstack/mastra @mastra/core zod

export TABSTACK_API_KEY="your-key-here"

@mastra/core (v1 or later) and zod are peer dependencies, so your app’s single instance of each is shared. tabstackTools is a named object keyed by tool name, ready to spread into an Agent:

import { Agent } from "@mastra/core/agent";
import { tabstackTools } from "@tabstack/mastra";

const agent = new Agent({
  id: "web-researcher",
  name: "Web Researcher",
  instructions: "Answer questions with current information, and always cite your sources.",
  model: "anthropic/claude-sonnet-4-6",
  tools: tabstackTools,
});

const result = await agent.generate("What are Vercel's pricing plans, with sources?");
console.log(result.text);

Want a subset? Every tool is exported individually:

import { Agent } from "@mastra/core/agent";
import { extractPageContentTool, researchQuestionTool, toolNames } from "@tabstack/mastra";

const agent = new Agent({
  id: "web-researcher",
  name: "Web Researcher",
  instructions: "Summarize pages and research questions.",
  model: "anthropic/claude-sonnet-4-6",
  tools: {
    [toolNames.researchQuestion]: researchQuestionTool,
    [toolNames.extractPageContent]: extractPageContentTool,
  },
});

For a custom key, base URL, or a shared client, build the tools explicitly:

import { createTabstackMastraTools } from "@tabstack/mastra";

const tools = createTabstackMastraTools({ apiKey: process.env.MY_KEY });
// or pass an SDK client you already have: createTabstackMastraTools({ client })

The tools

The model fills the inputs in, but the shapes are worth knowing: extract_structured_data takes url and json_schema_json (a JSON-encoded JSON Schema string), extract_page_content takes url, research_question takes query, generate_structured_data takes url, instructions, and json_schema_json, and automate_browser_task takes task plus optional url, guardrails, data, country, max_iterations, and max_validation_attempts.

Optional inputs

The model can pass these for finer control, and they are sent to Tabstack only when present:

• extract_structured_data, extract_page_content, generate_structured_data: effort ("min", "standard", or "max", where "max" suits JS-heavy pages), nocache to bypass the cache, and country as an ISO 3166-1 alpha-2 code for geotargeted fetches.
• research_question: mode ("fast" or "balanced") and nocache.

Good to know

• automate_browser_task runs non-interactively. It does not pause for human-in-the-loop form input, so it does not wait on a person. It returns the final answer plus the data it extracted and the pages it visited.
• Failed calls throw TabstackToolError, a normalized message plus an HTTP status for API errors. Mastra surfaces this in the tool result.
• Tool names, descriptions, and inputs match the @tabstack/langchain and Python langchain-tabstack packages, so inputs are the same across frameworks and languages.

Common use cases

• Give a Mastra agent cited, multi-source answers from the live web.
• Pull structured fields off a page into a Zod-defined shape.
• Geotarget a fetch by country to see region-specific pricing or availability.
• Keep tool names consistent across your TypeScript and Python agents.

Next steps

• Adapter on GitHub
• @tabstack/mastra on npm
• Mastra docs
• Get an API key