Docusign MCP is coming soon to GA!

GA officially launches on September 30! 

We’re excited to announce that we’re opening our Model Context Protocol (MCP) Server to every AI agent. In the coming weeks the Docusign MCP server will become generally available. Agreement intelligence and governed action, powered by our AI engine, Docusign Iris, will be callable from Claude, ChatGPT, Gemini, Copilot, Slack, and any MCP client, through a single standard interface.

The Docusign MCP Server enables teams to use Docusign’s Intelligent Agreement Management (IAM) platform from inside the familiar chat tools and third-party agentic systems that they’re already using – and use IAM to gain access to external agentic systems without leaving IAM. 

What this means for developers: 

Instead of writing custom API wrappers, parsing JSON payloads, and manually orchestrating endpoint calls, developers can expose Docusign’s Lifecycle Management and eSignature capabilities directly to AI agents.

1. Accelerate agentic engineering

• Out-of-the-box tooling: Eliminate thousands of lines of boilerplate code previously required for REST API integrations, webhooks, and custom function definitions.
• Pre-built capabilities: Connect AI models to pre-configured Docusign tools for contract analysis, envelope creation, and status polling without building function integrations from scratch.

2. Expanded surface availability & write-once portability

• Universal compatibility: Because MCP is an open, vendor-agnostic standard, tools written for the Docusign MCP Server work natively across Claude Desktop, Perplexity, Claude for Legal/SMB, ChatGPT, Copilot Studio, or custom-built internal agents.
• Cross-client execution: Build a single integration strategy and deploy it seamlessly across different AI platforms without re-architecting your backend.

3. Programmatic agreement execution & embedded UI

• Interactive experience cards: Enable AI agents to dynamically generate interactive MCP Cards directly inside client interfaces (e.g., in chat windows or legal IDEs) for agreement creation and template selection.
• Actionable lifecycle management: Let agents monitor real-time recipient status, flag pending actions, send nudges, and execute envelopes programmatically.

4. Agreement Intelligence (IAM + CLM)

• Context-aware queries: Give agents read access to search and parse structured/unstructured contract data, clauses, metadata, and historical context across Docusign Intelligent Agreement Management (IAM) and Contract Lifecycle Management (CLM).
• Natural language data extraction: Query complex agreement portfolios using natural language (e.g., "Find all active NDAs expiring next quarter with auto-renewal clauses").

5. Autonomous workflow execution

• End-to-end automation: Trigger and track complex Docusign Workflow Builder (Maestro) and CLM automated flows directly from external AI agents.
• Event-driven chains: Allow AI agents to analyze a document, select the appropriate workflow template, populate variables, and kick off approval routing with zero manual overhead.

6. Native enterprise governance & security

• Enforced OAuth & RBAC: Agents operate strictly within the boundaries of Docusign’s established Role-Based Access Control.
• User-level security: An agent authenticated via OAuth can only inspect, search, or execute actions on envelopes and documents that the authenticated user explicitly has permissions to view.

No need to wait for GA, the Docusign MCP Server is already available in beta! To learn more, you can view our documentation and ask questions on our community Q&A post.

For a real-world example, watch how Leon van Zyl built an AI-native agreement management agent using the Docusign MCP and the Claude Agent SDK: