How to connect PandaDoc to an AI agent: what GTM AI agents can do with PandaDoc in Day AI
Updated
The short answer. To connect PandaDoc to an AI agent, sign the agent in to PandaDoc's MCP server as yourself, so GTM AI agents, AI for GTM put to work on real deals, can turn what the buyer said on the call into a proposal, send it for e-signature and tell you where it stands. In Day AI, each rep clicks Connect on PandaDoc under Settings > Connectors and signs in, with no workspace setup first, and their agent uses PandaDoc's MCP server to create, send, e-sign and track documents and templates.

Let's say a scoping call for your marketing-analytics company just ended, a little before noon on a Tuesday, and you're the rep: the buyer wants two teams onboarded first, a pilot that starts in January, and the proposal in their inbox before their 2:00 internal review. You open Day AI and ask your agent to build the proposal from the call. It reads the meeting and last week's email thread, opens your standard proposal template in PandaDoc, and fills in the company, the two teams, the January start and the three people who need to see it. It shows you the draft and an approval card to create the document. You fix one team's name, approve, then approve the send. At 1:50 you ask whether they've opened it. The agent checks PandaDoc: the status says viewed, not signed yet. Your proposal reached the review with the buyer's own words in it.
Where the usual answer goes wrong. Most proposal automation maps CRM fields into a template's tokens. That fills in the company name and leaves the scope, the start date and the stakeholders to a rep retyping notes, because those things were said on a call, not entered in a field. Day AI builds the proposal from what was said on the deal and uses PandaDoc for the document, the signature and the status, with the rep approving what goes out. Your templates and your pricing approvals stay exactly as they are. What changes is where the words inside them come from.
The listen, draft, send model for PandaDoc, with Day AI
A proposal is a summary of a conversation with a signature line on it. Build it in that order.
- Listen: the scope comes from the call. Whatever the buyer asked for is already in the meeting and the follow-up email. In Day AI, your agent reads every call, email and meeting on the deal before it touches a template.
- Draft: the template does the formatting. Your PandaDoc templates already carry your terms and layout. In Day AI, the agent creates the document from the template you name and shows you what it filled in before anything is created.
- Send: the rep approves, then tracks. A proposal reaches the buyer only when a person says so. In Day AI, creating and sending a PandaDoc document wait as approval cards in the chat, and afterwards you ask the agent where the document stands.
How to connect PandaDoc to Day AI
Step 1: Find PandaDoc in Day AI's Connectors settings
Open Settings > Connectors and search for PandaDoc, or find it in Browse connectors. PandaDoc doesn't need a workspace Owner to save anything first, so each rep can connect on their own.
Step 2: Connect your PandaDoc account in Day AI
Click Connect and sign in to PandaDoc. Day AI loads the tools PandaDoc's MCP server offers for documents and templates. Your agent works with your own PandaDoc access, and a teammate connects their own account for theirs. You can also ask your agent "connect PandaDoc" in chat and it sends you the sign-in link.
Step 3: Decide what your Day AI agent may do in PandaDoc
Open PandaDoc's Tool permissions. Looking up templates and document status starts at Always allow. Creating, sending and other changes start at Needs approval, so each one appears as an approval card in the chat with Approve and Deny. Leave sending at Needs approval even after the team trusts the drafts: it's the step the buyer sees.
Step 4: Ask your Day AI agent for the proposal
Right after the call, ask in chat. "Draft the proposal for Northwind from today's call using our pilot template" builds the document for your approval. "Include their CFO from Thursday's email as a signer" sets the recipients before it's created. "Which of my proposals sent this month haven't been signed?" gives you the follow-up list. When a proposal is signed, ask the agent to move the deal forward in Day AI, or create a task for the kickoff.
Step 5: Give a Day AI skill the proposal report
For a Friday look at every open proposal, ask your agent to create a skill and pick PandaDoc in its Connectors picker. The skill works through its agent owner's PandaDoc sign-in. A skill's PandaDoc calls run without pausing for you, so give it the status report and keep creating and sending in chat. To see which other tools can feed a proposal, browse every Day AI integration and connector.
Where homemade proposal automation breaks
- Tokens fill the easy parts. A field-mapped template knows the company name and not the scope. In Day AI, the agent fills the template from what the buyer said on the call and in email.
- The rep retypes the notes anyway. The real work stays manual, so proposals slip a day. In Day AI, the draft is ready as soon as the rep asks after the call, to check and approve.
- Nobody checks the status. A busy rep doesn't look at a sent proposal until the deal goes cold. In Day AI, you ask where each proposal stands and get PandaDoc's current status.
- New recipients mean starting over. A late stakeholder breaks the automation. In Day AI, you name the new stakeholder and the agent proposes a fresh document from the same template for your approval.