← mai.deals

For Agents

Partner beta for connecting approved AI agents to MAI Deal Intelligence. Query M&A signals, discover targets, match buyers, and generate deal theses through MCP or A2A workflows.

Partner Beta MCP v1.27 A2A v1.0
25+
Signal Dimensions
93.3%
Internal Backtest
10M+
Private Profiles
9
MCP Tools
44d
Median Lead Time

Quick Start: MCP

Approved partner beta clients can add MAI to Claude Desktop or any MCP-compatible client:

{
  "mcpServers": {
    "mai-deals": {
      "url": "https://mcp.mai.deals/mcp",
      "transport": "streamable-http"
    }
  }
}

Once approved and configured, your agent can call the tools below directly.

Quick Start: A2A

Agent Card (discovery):

GET https://mai.deals/.well-known/agent-card.json

Send a query via JSON-RPC:

# Python example
import httpx, json

resp = httpx.post("https://api.mai.deals/a2a", json={
    "jsonrpc": "2.0",
    "id": "1",
    "method": "SendMessage",
    "params": {
        "message": {
            "messageId": "msg-001",
            "role": "user",
            "parts": [{
                "data": {
                    "skill_id": "discover_targets",
                    "params": {
                        "industry": "manufacturing",
                        "state": "TX",
                        "min_score": 60
                    }
                },
                "mediaType": "application/json"
            }]
        }
    }
})

task = resp.json()["result"]["task"]
print(json.dumps(task["artifacts"], indent=2))

Or use natural language (skill is auto-detected):

# curl example
curl -X POST https://api.mai.deals/a2a \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "id": "1",
    "method": "SendMessage",
    "params": {
      "message": {
        "messageId": "msg-001",
        "role": "user",
        "parts": [{"text": "Find manufacturing companies in Texas showing sell signals"}]
      }
    }
  }'

Available Tools

discover_targets
Find private companies showing sell-intent signals. Filter by industry, revenue, state, signal type.
find_buyers
Match a target company with ranked potential acquirers using 20-dimension matching engine.
get_signal_history
Signal timeline for a company, organized across MAI's proprietary signal dimensions.
generate_deal_thesis
Data-driven deal thesis from signal profile. Catalyst, rationale, risks, next steps.
check_signals
M&A signal status for a public company. Score, tier, signal count, trend.
company_profile
Full company profile: industry, financials, M&A assessment, competitive edge.
sector_heatmap
Industry M&A activity: signal density, tier distribution, top signal types.
structure_scenarios
Transaction structure analysis with capital stack, timeline, and deal killer checks.
request_engagement
Request full MAI engagement: matching, analysis, valuation, coordinated outreach.

Data Coverage

Authentication

MCP and A2A workflows are in approved partner beta. Access may require scoped credentials, use-case review, and workflow-specific limits.

Get Early Access

Integrate MAI into your deal sourcing workflow through an approved partner beta.

Request Partner Access

Protocols

MCP (Model Context Protocol)

Agent-to-tool interface. Your Claude, GPT, or any MCP-compatible agent calls MAI tools directly. Structured inputs, structured outputs. No prompt engineering needed.

A2A (Agent-to-Agent Protocol)

Agent-to-agent interface. Your agent discovers MAI via the Agent Card, sends tasks via JSON-RPC, receives structured results. Follows the A2A v1.0 specification under the Linux Foundation's Agentic AI Foundation.

MAI Deal Inc. — AI-powered M&A deal origination for the US lower middle market.