Datacircle MCP Integration
Guide for finding the best GTM APIs with PraisonAI agents using Datacircle's MCP server
Add GTM API Plans to AI Agent
Datacircle benchmarks go-to-market (GTM) APIs, such as email finders and enrichment APIs, and scores their plans. Its MCP server is public and needs no API key: https://api.datacircle.dev/mcp (Streamable HTTP).
Quick Start
- Install Dependencies
Install PraisonAI with MCP support:
bash pip install "praisonaiagents[mcp]" - Set Your Model's Key
Datacircle needs no key. Set your model’s API key:
bash export OPENAI_API_KEY=your_openai_api_key_here - Create a file
Create
datacircle_agent.pywith this code:python from praisonaiagents import Agent, MCP datacircle_tools = MCP("https://api.datacircle.dev/mcp") api_agent = Agent( instructions="""You help people pick a GTM API. Use list_apis for the APIs on Datacircle's sheet, and search_plans for their plans, best first.""", model="gpt-4o-mini", tools=datacircle_tools, ) api_agent.start( "Which 3 Employee email API plans score highest on Datacircle? " "Give each one's provider, score and price per 1,000." ) - Run the Agent
Run the script:
bash python datacircle_agent.pyThe agent calls
search_plansand answers with the 3 best Employee email plans.
Tools
| Tool | What the tool does |
|---|---|
list_apis | Lists the GTM APIs on Datacircle, with the number of plans and providers for each API |
search_plans | Finds the plans of one API or of every API. You can search by provider or plan name. You can sort by score or by price |
Each plan comes with a score out of 100, a price per 1,000 and a page on datacircle.dev. Both tools are read-only. Each field of a plan is explained at datacircle.dev/fields.