TL;DR

[Benchmark] Tavily 2-step Research is a marketing workflow that chains Tavily to automate a common task. Internal benchmark-only recipe: Tavily search + answer for a topic, 2 tool calls. Once configured, it saves n/a and runs through Claude Code, Cursor, Windsurf or any MCP-compatible AI agent.

🔍

MarketingBeginner

[Benchmark] Tavily 2-step Research

Internal benchmark-only recipe: Tavily search + answer for a topic, 2 tool calls.

10-30s per run1 MCP requiredSaves n/a

How it works

🔍Tavily

Automated

1tavily.search(broad query)2tavily.extract(top URL) for details

Hostable — runs in your browser1/1 MCPs hosted

Run with MCPizy

New

Execute this recipe in your browser — no local install, no Claude Code. Streams results live.

Whitelisted MCPs: perplexity, notion, anthropic, openai, tavily, firecrawl, coingecko, stripe, slack, github, gitlab, linear, resend, sendgrid, elevenlabs, shopify, sentry, posthog, supabase-mcp, context7, deepwiki~4k tokens · ~$0.012 est.

Why this combo?

Benchmark only — 2-step happy-path for APC.

Without this workflow

N/A

With MCPizy

N/A

Business value

Concrete ROI — not marketing fluff.

Time saved

n/a

Benchmark only

Workflow steps

1
tavily.search(broad query)
2
tavily.extract(top URL) for details

Use cases

APC benchmark: 2-tool-call scenario

MCPs required

🔍

Tavily

Tavily MCP Server

View

Agent prompt (copy into Claude Code)

This prompt is the workflow. Paste into Claude Code, Cursor, or Windsurf.

You are a research agent. Given a topic:
1. Call tavily.tavily-search with query=<topic> and max_results=3
2. Pick the most relevant URL from step 1
3. Call tavily.tavily-extract with urls=[that_url]
4. Respond with a 3-sentence summary combining both calls.
Do not call any other tool.

Trigger & credentials

How this workflow fires and what env vars you need.

.env.example

ManualTrigger

🔍Tavily· 1 var

TAVILY_API_KEYGet key

Tavily API key

e.g. tvly-...

One-command deploy

Install everything — MCPs, prompt, env template — in a single call.

$ mcpizy recipe install bench-tavily-research-2

✓ Installs all 1 MCP server
✓ Writes prompt to ~/.mcpizy/prompts/bench-tavily-research-2.md
✓ Generates .env.example in current directory
✓ Ready to paste into Claude Code

Requires mcpizy CLI v1.1+ — install via npm i -g mcpizy.

Quick install (MCPs only)

10-30s per run

$ n/a

More Marketing recipes

🔮📝

Competitor Watch Automation

Run daily Perplexity searches on competitors and log product updates, pricing changes, and news to a Notion tracker.

🔥💬

Product Hunt Scraper Alert

Scrape Product Hunt daily with Firecrawl and send trending posts in your category to Slack so your team never misses a launch.

🔍

[Benchmark] Tavily News Search

Internal benchmark-only recipe: one Tavily search call, returns a short summary. Not intended for production use.

🔍

[Benchmark] Tavily 3-query Multi-angle

Internal benchmark-only recipe: 3 Tavily searches on different angles of a topic, then synthesise.

Frequently asked questions

What is this workflow?

[Benchmark] Tavily 2-step Research is a marketing automation that uses Tavily together via the Model Context Protocol. Internal benchmark-only recipe: Tavily search + answer for a topic, 2 tool calls.

How long does setup take?

Setup takes around 10-30s per run. You install the required MCP servers with `n/a`, connect your accounts, and the workflow is ready to run.

How much time does this workflow save?

Once running, this workflow saves n/a. The concrete business value: Benchmark only.

Which MCP servers do I need for this?

You need 1 MCP server: Tavily (mcpizy install tavily). All are installable in one command via the MCPizy CLI and configured in your `.claude.json` or `.cursor/mcp.json`.

Does this work with Claude Code, Cursor, and Windsurf?

Yes. The workflow runs with any MCP-compatible AI agent — Claude Code, Claude Desktop, Cursor, Windsurf, VS Code with Copilot, and custom agents built on the MCP SDK. The MCP servers are identical across clients; only the config file path (`.claude.json` vs `.cursor/mcp.json`) changes.

Start building this workflow

Install the required MCPs from the marketplace and automate this in 10-30s per run.

$ n/a

🔍Install Tavily

Free to install. Connect your accounts and this workflow runs itself.

You are a research agent. Given a topic: 1. Call tavily.tavily-search with query=<topic> and max_results=3 2. Pick the most relevant URL from step 1 3. Call tavily.tavily-extract with urls=[that_url] 4. Respond with a 3-sentence summary combining both calls. Do not call any other tool.

$ mcpizy recipe install bench-tavily-research-2 ✓ Installs all 1 MCP server ✓ Writes prompt to ~/.mcpizy/prompts/bench-tavily-research-2.md ✓ Generates .env.example in current directory ✓ Ready to paste into Claude Code