Your team already works alongside AI assistants. Now those assistants can work inside LendAPI. LendMCP is our Model Context Protocol server: connect Claude, ChatGPT or any MCP client once, and it can build and test across every LendAPI module, from Product Studio and the Pricing Engine to Workflow Builder, Contracts, Documents, Communications and Labs. You decide what goes live. We’re rolling LendMCP out to all LendAPI users in batches, starting now.

LendMCP: Your AI Agent, Inside LendAPI
One connection, every module
Connect once
Add LendAPI as a connector in claude.ai, Claude Desktop, Claude Code, ChatGPT developer mode or any other MCP-capable tool. Sign in with your LendAPI account, approve access on the consent screen, and you’re connected. There are no API keys to copy and paste.
One approval covers both Live and Sandbox, and every tool call says which environment it runs in. Super Admins, Tenant Admins and Managers can approve a connection. It appears in Settings > API Keys, where you can revoke it at any time.
Build across every module
LendMCP now covers 166 tools. Describe the product you want in plain language, and your agent can do the work end to end:
Product Studio: create a product, write its pages, routing, settings and theme, then validate it.
Pricing Engine: build an offer table, test it against sample values and publish a version.
Workflow Builder: draft the flows that run your decision trees, pricing and communications, then validate and test-run them.
Decision trees: create variables and outcomes and build rules node by node.
Contracts: upload a PDF and turn its sample values into fillable fields, or author HTML templates.
Documents and communications: set up document requests, library documents, and email and SMS templates.
Labs: configure A/B tests, Shadow tests and Backtests, then see which applications champion and challenger disagreed on.
Your agent can also review applications: read an application’s stage, provider results, decisions and documents, add a note or create a manual offer.
See the whole product before you ship it
Ask for a bird’s-eye view, and the agent draws the product as an interactive flow map right inside claude.ai or Claude Desktop: every step, its widgets and providers, and how each outcome routes. It can audit a draft for configuration warnings, compare two versions and tell you what publishing would change.
You stay in charge
Work lands as drafts for your team to review, and every change is kept in version history. Build and test in Sandbox, then copy the product to Live when you’re ready. In Product Studio, anything that would reach live applicants asks you to confirm before it happens.
The practical impact: a product that used to take days of clicking through modules can be drafted in one conversation, and your team spends its time reviewing and deciding instead of configuring.
Rolling out now
We’re rolling LendMCP out to all LendAPI users in batches over the coming weeks. When it reaches your tenant, an admin can connect it straight from your AI client. Want it sooner, or have a workflow you’d like your agent to handle? Talk to your customer success manager.












