---
title: "AI Agents"
description: "Connect AI agents to SignatureAPI through the hosted MCP server, give coding agents the docs and skills, and keep live sends under human approval."
canonical: "https://signatureapi.com/docs/ai-toolkit/agents"
source: "https://signatureapi.com/docs/ai-toolkit/agents.md"
last_verified_api_version: "1.18.0"
last_modified: "2026-11-18T13:19:17Z"
---

# AI Agents

Connect AI agents to SignatureAPI through the hosted MCP server, give coding agents the docs and skills, and keep live sends under human approval.

_Last verified against API version 1.18.0._

SignatureAPI is built to work alongside AI. Whether you are connecting an off-the-shelf agent through MCP, feeding our documentation into an LLM at build time, or designing an end-to-end agentic signing experience, this section is your starting point.

## Connect an agent through MCP

Our hosted [Model Context Protocol](https://modelcontextprotocol.io/introduction) server lets any MCP-compatible client send envelopes, manage signing, and look up documentation on behalf of an account. Sign in once through OAuth and the agent is ready to act.

The fastest way to connect a coding agent is to let it do the work. Paste this into Claude Code, Codex, Cursor, or any agent that can read a URL:

```text
Set up SignatureAPI for me. Read https://signatureapi.com/docs/ai-toolkit/mcp/connecting-clients.md
and follow the section for the agent you are running in. If I am working on a codebase, install the
SignatureAPI plugin (skills plus MCP server) rather than the MCP server alone. Ask me before opening
a browser for sign-in, and never print tokens or keys. When you are done, prove it works by listing
my 5 most recent test-mode envelopes, then tell me what you changed and how to undo it.
```

Prefer to do it yourself? [Connecting clients](https://signatureapi.com/docs/ai-toolkit/mcp/connecting-clients) has a one-command installer, one-click buttons for Cursor and VS Code, and manual steps for every supported app.

- [MCP server overview](https://signatureapi.com/docs/ai-toolkit/mcp/overview): What the server is, the endpoint, transport, and when to use it instead of the REST API.

- [Connecting clients](https://signatureapi.com/docs/ai-toolkit/mcp/connecting-clients): Configure Claude Desktop, Claude.ai, ChatGPT, Cursor, Claude Code, Codex, and more.

- [Authorization](https://signatureapi.com/docs/ai-toolkit/mcp/authorization): OAuth 2.1 with Dynamic Client Registration and discovery. For your own users' agents, build on the REST API instead.

- [Tools and resources](https://signatureapi.com/docs/ai-toolkit/mcp/tools): Browse the tools available to agents, including the upload flow and the envelope resource.

- [Managing connections](https://signatureapi.com/docs/ai-toolkit/mcp/managing-connections): Review, rename, and revoke connected applications from the dashboard.

## Guides

- [Send contracts with an agent](https://signatureapi.com/docs/ai-toolkit/guides/agent-sends-contract): Connect Claude, ChatGPT, or Cursor, check a draft in test mode, and send it once you approve.

- [Send an NDA to a list](https://signatureapi.com/docs/ai-toolkit/recipes/nda-batch): One prompt that sends an NDA to several people and reports who has signed.

- [Build with coding agents](https://signatureapi.com/docs/ai-toolkit/guides/build-with-coding-agents): Use the agent skills in Claude Code, Codex, or Cursor to design, build, and test your integration.

- [Agent safety](https://signatureapi.com/docs/ai-toolkit/guides/agent-safety): Test mode by default, human approval before live sends, and scoped access.

## Develop with agents

When you are the developer writing code against SignatureAPI, feed the documentation to your AI coding assistant so it generates accurate integration code.

- [Agent skills and plugin](https://signatureapi.com/docs/ai-toolkit/develop/agent-skills): Four skills that design, build, and diagnose e-signature integrations and answer questions from the docs. Install them alone or as a plugin with the MCP server.

- [llms.txt](https://signatureapi.com/docs/ai-toolkit/develop/llms-txt): Discoverable indexes of the SignatureAPI documentation, designed to be consumed by LLMs at build or prompt time.

- [Context7](https://signatureapi.com/docs/ai-toolkit/develop/context7): Pull up-to-date SignatureAPI documentation into Cursor, Claude Code, and other MCP-aware editors via Context7.

## Use cases

See what becomes possible when an agent drives the entire signing flow.

- [Agentic ceremonies](https://signatureapi.com/docs/ai-toolkit/agentic-ceremonies): A showcase of fully agent-driven signing experiences built on top of SignatureAPI.
