Start and fill intake
Create a fill session, answer the intake conversationally, and generate the filled application PDFs.
The Bindly MCP connector is a Model Context Protocol server for your workspace. Connect it to an AI tool and the assistant can start and fill intake, summarize risk, and submit to Hedge, using natural language.
Start and fill intake
Create a fill session, answer the intake conversationally, and generate the filled application PDFs.
Extract from PDFs
Pull details out of an uploaded PDF, such as a prior application or a loss run, into the session.
Summarize risk
Read an underwriter-facing summary of the risk flags on a session’s collected intake.
Submit to Hedge
Submit a ready session to Hedge on your workspace’s connected account, and request a Hedge appointment.
The connector exposes both read tools (list sessions, read a session, risk summary) and action tools (create a session, answer intake, extract, fill forms, submit to Hedge). Read tools are marked read-only so a client can surface which tools change data.
The connector works with any MCP-capable client, including Claude, ChatGPT, Cursor, and Claude Code.
Copy the connector URL.
https://engine.bindly.insure/mcpAdd it to your AI tool. Paste the URL into your client’s connector or MCP server settings:
Sign in and approve. The tool opens Bindly. Sign in and approve the connection. Access is scoped to your workspace.
Work in natural language. The assistant can now do things like “start an intake for Acme HVAC in Texas for general liability”, “pull the details out of this PDF”, or “submit that session to Hedge”.
The connector authenticates with OAuth 2.1, the same model the CLI uses. A connected tool stays authorized until its token expires or you sign it out. Approve connectors only in tools you trust, since the Bindly MCP can act on your workspace. See Authentication for the full model.