Your AI agent signing documents in minutes
Connect Claude, ChatGPT, Cursor, Codex, or Gemini to ZapSign and create documents, send them for signature, and track status — by natural language or API.
https://mcp.zapsign.com.br/mcp // Paste this URL in Settings → MCP / Connectors // Log in with your ZapSign token and you're done. // Ask: "send this contract to // someone@company.com to sign"
What do you want to do?
Three paths, from zero to your first signed document. Choose the one that fits your case.
Use ZapSign from inside my AI
Connect your assistant (Claude, ChatGPT, Cursor…) via MCP and manage documents through conversation. No coding required.
See tutorials by tool →Integrate signatures into my product
Use the REST API to create documents, define signers and receive status webhooks. First document in 10 minutes.
API Quickstart →Give signing powers to my agent
Run the ZapSign open-source MCP server (locally or in production via SSE) and expose 25+ tools to your autonomous agent.
MCP Server guide →Choose between MCP and REST API
Compare when to use the MCP server (natural language, typed tools) versus direct HTTP integration in your application.
MCP vs REST API →Connect your favorite tool
Tested step-by-step guides for each AI client, always with ZapSign’s official MCP server.
Claude
Paste the MCP URL in Claude Desktop → Connectors. The fastest path — no Node.js install.
ChatGPT
Remote MCP connector via URL in ChatGPT developer mode. Nothing to host.
Cursor
MCP in the editor: add the URL in Settings → MCP and create real documents in the agent loop.
Codex CLI
Configure ZapSign MCP in Codex config.toml and automate documents from the terminal.
Gemini CLI
Add the MCP server to Gemini CLI settings.json and use ZapSign tools in chat.
REST API
No MCP: create your first document with a POST and receive status via webhook.
What AI agents already do with ZapSign
Practical prompt examples and flows for e-signature with AI.
ChatGPT signing contracts
Send contracts for signature from ChatGPT via the MCP connector.
See example →HR onboarding with AI
Automate employment contracts and HR terms with batch signers.
See example →Rental contract by agent
Generate and send lease agreements from templates with variables.
See example →Automatic sales proposal
Create commercial proposals and collect the client signature in minutes.
See example →From prompt to signature in 3 steps
Generate your API token
For testing, create an account on the ZapSign sandbox and copy the api_token from Settings → Integrations → API. For production, use app.zapsign.com.br.
Connect the MCP server
Go to Settings → MCP / Connectors in your AI client and add the URL mcp.zapsign.com.br/mcp. The server will ask for your token — no file editing needed.
Chat with your documents
Ask in natural language: "create a contract from template X and send it to maria@company.com to sign". The agent does the rest.
What your agent learns to do
The MCP server exposes the main ZapSign API operations as typed tools.
📄 Documents
Create by upload or template, list, detail, update, delete, attach extra document and position signatures.
✍️ Signers
Add, update and remove signers, check status and sign in batch.
📋 Templates
List and use templates with dynamic variables to generate standardized contracts.
🔍 Background checks
Individual and company verifications with status tracking.
🔔 Webhooks
Create and remove webhooks to receive events for signed, refused documents and more.
🕐 Timestamp
Add a timestamp to documents for additional legal evidence.
Documentation your AI reads on its own
ZapSign publishes resources designed for LLM consumption — point your agent at them and reduce endpoint hallucination.
Documentation llms.txt
Complete API index in llms.txt format: authentication, documents, signers, templates, webhooks and SDKs.
docs.zapsign.com.br/llms.txt →Docs with ?ask= response
Ask in natural language: GET docs.zapsign.com.br/master.md?ask=how+to+create+document returns the right doc section.
This site’s llms.txt
This hub is also agent-indexable: structure and summary of all tutorials in /llms.txt.
llms-full.txt (full text)
Expanded version with the full content of every sitemap page — ideal for RAG and LLM citation.
agents.zapsign.com.br/llms-full.txt →Frequently asked questions
What is the ZapSign MCP server?
It is an open-source Model Context Protocol server that exposes the ZapSign API as tools for AI agents. With it, assistants like Claude, Cursor, Codex and Gemini create documents, add signers, use templates and manage webhooks via natural language. Code at github.com/ZapSign/api-mcp; npm package mcp-server-zapsign.
Do I need to pay to test the integration?
No. ZapSign has a free plan with 5 documents per month, and the API can be tested with your account token. The MCP server is open source and free.
Where do I find my API token?
For testing, use sandbox.zapsign.com.br → Settings → Integrations → API. For production, go to app.zapsign.com.br. Use it in the header Authorization: Bearer YOUR_TOKEN or in the ZAPSIGN_API_KEY MCP variable.
Is ZapSign documentation optimized for AI?
Yes. In addition to the llms.txt, the docs respond to natural language queries via GET docs.zapsign.com.br/master.md?ask=your-question — ideal for RAG and autonomous agents.
Does an AI-triggered signature have legal validity?
Validity comes from the platform, not who triggers the request. Documents created via API or MCP follow the same ZapSign signature flow, with a full audit trail. ZapSign is ISO 27001 certified, with over 70 million documents signed in 81+ countries.
Is my data safe when connecting an AI agent to ZapSign?
Yes. ZapSign is ISO 27001 certified, with over 70 million documents signed in 81+ countries. Your api_token stays in the MCP server or client configuration — it is never sent to the language model. Documents created via agents follow the same signature flow with a full audit trail.
Ready to give superpowers to your agent?
Create your free account, get your token and connect your AI in less than 10 minutes.