Fenceline MCP Integration
Imagine never having to manually file a permit, chase down a vendor quote, or update your books again. With MCP, AI doesn't just understand your business — it runs it.
Real-time streaming for web browsers
GET/POST /mcp/sse
Simple request/response for servers
POST /mcp/rpc
Direct integration with Claude Desktop
stdio transport
Real-time bidirectional communication
SHA-256 hashed keys with contractor-scoped access and MCP scope validation
Complete contractor segregation - access only your data
Complete logging of all operations for compliance and security
Connect your applications directly to Fenceline's MCP server:
curl -X POST \
-H "Content-Type: application/json" \
-H "X-API-Key: YOUR_API_KEY" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}' \
https://mcp.fenceline.ai/mcp/rpc
curl https://mcp.fenceline.ai/mcp/info
Returns available transports and examples
Your API key must include one of: mcp:tools
, mcp:all
, admin
, or read:all
AI-powered search across your projects, estimates, materials, and business data
Automated permit applications and municipal portal interactions
Connect to supplier systems for pricing, ordering, and inventory management
Integrate with QuickBooks, Xero, and other financial platforms
Connect to Salesforce, HubSpot, Slack, and communication tools
Generate beautiful, customizable contracts with integrated signing workflows
While everyone else is building chatbots, we're building AI that actually works in your business. MCP is the missing piece that turns conversations into actions.
AI navigates permit portals, files applications, schedules inspections — even on websites from 1995
QuickBooks, Salesforce, DocuSign — AI handles the tedious work while you focus on growing your business
AI shops around, compares prices, places orders, tracks deliveries — getting you the best deals automatically
The future of business isn't about better software — it's about AI that uses all your software for you.
Model Context Protocol is the open standard that makes this possible. Our implementation shows what's achievable when AI can actually interact with business systems.