The Model Context Protocol (MCP) is an open standard that allows AI models to securely connect to external tools. You can think of it as a universal interface that lets AI systems move beyond generating text and begin performing real-world actions.

Is this ready for my whole team?

Version 1 is released as an open-source project on GitHub. It is primarily intended for developers and technically confident users who are comfortable configuring their own AI environments.

Connection instructions are available within your AI environment (for example Claude), which treats the Scrive MCP server as a secure connector.

Are there rate limits?

There are no additional AI-specific limits. The only limits applied are those associated with your Scrive subscription, such as document volume or SMS credits.

What would I use this for?

Different departments could use this for a variety of things, including preparing agreements from templates, setting up automatic reminders or finding specific documents, all automated with minimal input required from the end-user. All they have to do is ask for what they need in their own words and our MCP translates their requests into action. Some departmental examples include: Sales “Prepare the Enterprise Agreement for TechCorp using our Q1 template. Set authentication to Swedish BankID and send it to their CEO.” HR “Show me which new hires have not signed their onboarding documents yet and send a reminder to those starting next Monday.” Legal “Find all PDFs in my local folder and create a document bundle for [Partner Name]. Start the signing process.”

Scrive MCP server (beta)

Agentify your agreement lifecycle

Connect AI agents to combine generative AI with verified identity and eIDAS-compliant signatures, bridging the trust gap and automating legally binding agreement workflows.

Join the waitlist

Sign-up and join the waitlist

Version 1 will be released as open source. Early adopters will soon be able to download the server directly from our GitHub and begin building. In this initial phase, the tool is intended for developers and technical teams rather than non-technical end users.

AI can generate documents – Scrive ensures they become trusted agreements

With the Scrive MCP server, AI agents can trigger real agreement workflows: preparing documents, verifying identities and collecting legally binding signatures. Sign-up and you will be the first to know when it is live!

Join the waitlist

Why trusted AI matters

Most AI tools can draft text but they cannot verify who is interacting with the system or ensure the outcome will hold up in court. Scrive MCP addresses this by bringing trusted identity and legally binding signing into AI-driven workflows.

ID-verified signatures
Use trusted eIDs such as Swedish BankID, MitID or Finnish BankID to authenticate signers directly from an AI-driven process.

Legally binding outcomes
Because MCP builds on Scrive’s core products, documents produced through these workflows meet strict eIDAS compliance standards.

Context-aware logic
The integration understands when to use different Scrive services, depending on the complexity and requirements of the signing process.

How it works

Following the approach used by trusted platforms such as Atlassian, the Scrive MCP server acts as a bridge between your AI tool and Scrive’s APIs. You do not configure the connection inside MCP itself but rather you add the Scrive connector within your preferred AI environment (for example Claude Desktop).

The workflow

1. The AI agent receives your natural language command

2. The MCP server translates that command into a secure Scrive API request

3. Scrive performs the action, such as searching documents, sending reminders or triggering signing workflows

Core capabilities

The Scrive MCP server provides a set of tools that allow AI agents to interact directly with your agreement workflows.

Deep search and filtering
Ask your AI to find documents by title, content, modification date or signing status.

Automated reminders
Tell the AI to “Remind the signee to sign the NDA” and it will trigger the reminder email immediately.

Document and bundle creation
Generate individual documents or complex bundles with multiple PDFs and signees through a single prompt.

Developer resources

Built for modern AI environments

The Scrive MCP server provides a standardised interface that can connect to AI environments that support MCP, including:

Claude Desktop
ChatGPT
Copilot

Start building trusted AI agreement workflows

To experiment with AI-driven agreements using Scrive’s open-source MCP server, join the waitlist for the managed beta.

Join the waitlist

FAQ

What is MCP?
The Model Context Protocol (MCP) is an open standard that allows AI models to securely connect to external tools. You can think of it as a universal interface that lets AI systems move beyond generating text and begin performing real-world actions.
Is this ready for my whole team?
Version 1 is released as an open-source project on GitHub. It is primarily intended for developers and technically confident users who are comfortable configuring their own AI environments.
How do I connect it?
Connection instructions are available within your AI environment (for example Claude), which treats the Scrive MCP server as a secure connector.

Are there rate limits?
There are no additional AI-specific limits. The only limits applied are those associated with your Scrive subscription, such as document volume or SMS credits.
What would I use this for?
Different departments could use this for a variety of things, including preparing agreements from templates, setting up automatic reminders or finding specific documents, all automated with minimal input required from the end-user. All they have to do is ask for what they need in their own words and our MCP translates their requests into action.

Some departmental examples include:

Sales
“Prepare the Enterprise Agreement for TechCorp using our Q1 template. Set authentication to Swedish BankID and send it to their CEO.”

HR
“Show me which new hires have not signed their onboarding documents yet and send a reminder to those starting next Monday.”

Legal
“Find all PDFs in my local folder and create a document bundle for [Partner Name]. Start the signing process.”

What are the core capabilities of Scrive's MCP?

Here are the available MCP Tools in the current early access version.

Tool	Description
`create_document`	Upload a PDF file from an absolute local path.
`list_documents`	List documents with filtering and sorting options.
`get_document`	Retrieve a document’s full JSON representation.
`add_party`	Add a signing party to an existing document.
`start_signing`	Start the signing flow for a document.
`remind_document`	Send reminders to signatories who have not signed yet.
`get_usage_stats`	Retrieve daily or monthly Scrive usage statistics.
`create_flow_draft`	Create a new Journey flow draft.
`add_document_to_draft`	Add a PDF document to an existing Journey draft.
`list_flow_drafts`	List Journey flow drafts.
`get_flow_draft`	Retrieve a Journey flow draft.
`delete_flow_draft`	Delete a Journey flow draft.
`start_flow`	Start a Journey flow draft.
`add_participant_to_draft`	Add a participant to a Journey draft and regenerate steps.

Tool

Description

create_document

Upload a PDF file from an absolute local path.

list_documents

List documents with filtering and sorting options.

get_document

Retrieve a document’s full JSON representation.

add_party

Add a signing party to an existing document.

start_signing

Start the signing flow for a document.

remind_document

Send reminders to signatories who have not signed yet.

get_usage_stats

Retrieve daily or monthly Scrive usage statistics.

create_flow_draft

Create a new Journey flow draft.

add_document_to_draft

Add a PDF document to an existing Journey draft.

list_flow_drafts

List Journey flow drafts.

get_flow_draft

Retrieve a Journey flow draft.

delete_flow_draft

Delete a Journey flow draft.

start_flow

Start a Journey flow draft.

add_participant_to_draft

Add a participant to a Journey draft and regenerate steps.