How is this different from connecting PandaDoc MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the PandaDoc MCP API changes?

We track the PandaDoc MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different PandaDoc tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in PandaDoc?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the PandaDoc MCP server myself?

No. TrustGate connects to the PandaDoc MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to PandaDoc directly.

Connect PandaDoc MCP to your AI agents, securely.

Q: Is PandaDoc MCP secure?

The server itself exposes 32 tools, and some of them can write to your PandaDoc account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches PandaDoc.

One secured endpoint for every PandaDoc call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try PandaDoc MCP now

What your agents can ask once PandaDoc MCP is connected.

Select the tools your AI agents can call

32 tools from the PandaDoc MCP schema.

Showing 32 tools

pandadocmcp_documents_listreadLists documents with filters for status, folder, tag, free-text search, sorting and created or completed date ranges, returning paginated results.
pandadocmcp_documents_searchreadFull-text search across documents matching content and metadata, with optional status, date-window and pagination filters.
pandadocmcp_documents_details_getreadRetrieves full details for a document including metadata, recipients, fields and status.
pandadocmcp_documents_content_getreadReturns the content of a document as plaintext or markdown by document ID.
pandadocmcp_documents_status_getreadReturns the current status of a document by ID.
pandadocmcp_documents_summary_getreadReturns an AI-generated or standard summary for a document at detailed, short or headline granularity.
pandadocmcp_documents_metadata_getreadReturns AI-extracted metadata fields from a document as structured key-value pairs, combining document and content data.
pandadocmcp_documents_metadata_batch_getreadRetrieves metadata for 1 to 40 documents in one request. A failure for one document does not fail the batch.
pandadocmcp_documents_audit_trail_getreadRetrieves the full audit trail for a document, showing all events including views, signatures and status changes.
pandadocmcp_documents_createwriteCreates a document from a PandaDoc template, from markdown content, or from a PDF or DOCX file URL. Creation is asynchronous, so the status must be polled until the document is Draft or Error.
pandadocmcp_documents_updatewriteUpdates a draft document's name, recipients, fields, tokens, images, tables, pricing tables, tags and metadata. The document must be in draft status.
pandadocmcp_documents_sendhigh riskSends a draft document to its recipients for review and signature, with optional subject, message, reply-to, sender, approver selection and forwarding settings. The silent flag disables email notifications for recipients and the sender and stops scheduled reminders, and skip_unfilled_variables should stay false unless the user has explicitly chosen to send with empty variables.
pandadocmcp_documents_status_changehigh riskManually changes a document's status. Only Completed, Paid, Expired and Declined are settable; every other status is managed automatically by PandaDoc. It can notify recipients of the change.
pandadocmcp_documents_archivehigh riskArchives a document by ID, removing it from active lists without permanently deleting it. Annotated destructive by the vendor.
pandadocmcp_documents_linkwriteAssociate a document with a CRM object such as a HubSpot deal, so the document appears in that CRM record.
pandadocmcp_documents_reminder_sendwriteSend a reminder to recipients of a document that is out for signature, so the ones who have not completed their part are prompted again.
pandadocmcp_files_request_upload_urlwriteCreate a short-lived upload slot for a PDF or DOCX file.
pandadocmcp_recipients_add_ccwriteAdds a CC, meaning non-signing, recipient to a document using an existing contact or contact group ID. It cannot add to expired or declined documents.
pandadocmcp_recipients_editwriteUpdates a recipient's email, name, phone, company, job title, address, delivery methods or post-signing redirect. A signer's email cannot be changed after they have signed.
pandadocmcp_recipients_reassignhigh riskReplaces a signer with another contact, transferring all of their assigned fields to the new signer. Recipients who have already signed cannot be reassigned.
pandadocmcp_recipients_deletehigh riskRemoves a recipient from a document. Signers can only be removed while the document is in draft; CC recipients can be removed in any status except expired or declined. Annotated destructive by the vendor.
pandadocmcp_contacts_createwriteCreate a contact in the workspace address book from an email address, and return it with its `id`, the contact id that `recipients_add_cc` and `recipients_reassign` take.
pandadocmcp_contacts_listreadResolve an email address to the contact it belongs to.
pandadocmcp_contacts_searchreadSearch the current workspace's contacts by first name, last name, or company.
pandadocmcp_templates_listreadLists templates with optional filters for search, tags, folder and shared or deleted status.
pandadocmcp_templates_details_getreadReturns full details for a template including its roles, fields, tokens and pricing tables. Pricing table column names are per-template, so this is the prerequisite for populating them.
pandadocmcp_templates_createwriteCreates a new template from a secure, publicly accessible PDF URL, with optional name, folder, metadata, custom tokens and owner.
pandadocmcp_templates_approval_workflow_updatewriteCreate or replace the Submit for approval configuration on a template.
pandadocmcp_templates_duplicatewriteCreate a copy of an existing template within the same workspace.
pandadocmcp_templates_updatewriteUpdate an existing template's CUSTOM variables (tokens) and/or its signer roles.
pandadocmcp_templates_deletehigh riskDelete a template by ID.
pandadocmcp_documents_fields_assignwriteAssigns, reassigns or unassigns document fields to recipients by mapping each field UUID to a recipient UUID, or null to unassign. The document must be in draft status.

What PandaDoc MCP exposes

PandaDoc MCP groups its 32 tools into 4 categories, each with a suggested starting policy.

All 32 tools as JSON, ready to paste.

Category	Covers	Access	Suggested policy
Documents	Listing documents with status, folder, tag, sorting and date-range filters, full-text search, reading content as plaintext or markdown, full details with recipients and fields, status, AI-generated summaries at three granularities, AI-extracted metadata singly or batched across up to 40 documents, the complete audit trail, and creating, updating, sending, archiving or manually restatusing a document	Read / Write	documents_send is the outbound act on this connection and its silent flag suppresses every notification, so log the flag as well as the call. documents_status_change can mark a contract Completed or Paid, which asserts something about the real world rather than the record
Recipients	Adding a CC recipient from an existing contact, editing a recipient's email, name, phone, company, address or post-signing redirect, replacing a signer with another contact and transferring their fields, and removing a recipient	Write	recipients_reassign changes who signs a contract and moves their fields across. A signer's email cannot change after they have signed and signers can only be removed while the document is in draft, which are useful built-in floors
Templates	Listing templates with search, tag, folder and shared or deleted filters, reading a template's full details including roles, fields, tokens and pricing tables, and creating a new template from a publicly accessible PDF URL	Read / Write	templates_details_get is the prerequisite for populating pricing tables correctly, since column names are per-template rather than fixed. templates_create fetches a remote PDF, so the URL is an ingest path worth logging
Fields	Assigning, reassigning or unassigning document fields to recipients, where each item maps a field UUID to a recipient UUID or null to unassign	Write	It only works while the document is in draft, which keeps field assignment away from anything already out for signature. Passing null unassigns rather than leaving a field alone, so an omission and a clear differ

Connect PandaDoc MCP to your AI agent

Four steps, under ten minutes.

01
Sign in to NeuralTrust
Open the NeuralTrust platform, sign in and go to Getting started, then Connect your tools.
02
Choose your agent
When asked which agent you use, pick Claude Code, Codex, Cursor, Windsurf, Copilot or your own code.
03
Connect PandaDoc
Find PandaDoc in the tool list and connect it.
04
Connect your agent
Follow the steps on screen to add the config to your agent, then send the test prompt.

Permissions down to the single tool

You do not allow PandaDoc MCP or block it. You allow the exact tools each agent can call, one by one.

PandaDoc MCP	32 TOOLS AVAILABLE
Code review agent	4 tools, read only
Issue triage bot	9 tools
Platform team assistant	all 32 tools

Tool by toolConnecting the server is not the unit of access. Each tool is granted on its own.
Default denyA tool you have not enabled is never exposed to the agent.
Approval on writesTools that change or delete things can require a human to confirm.

Is PandaDoc MCP secure?

One endpoint instead of raw credentials. Every call is checked against policy and logged.

AI agentsConnect through your MCP client configuration.	NeuralTrustApply controls based on your configuration.	PandaDoc MCP32 tools on PandaDoc's own MCP server, routed through Scalekit AgentKit over OAuth 2.1 with dynamic client registration, which stores and refreshes each user's tokens so the agent never handles credentials. PandaDoc labels them 11 read, 8 write and 3 destructive, and every tool also publishes an idempotency flag indicating whether a retry is safe.
Coding assistants	Connection & credentialsper user	Documents17 tools
Internal copilots	Tool access policiesdeny by default	Recipients7 tools
Workflow automations	Request & response inspectioninline	Templates7 tools
Compatible MCP clients	Activity monitoringalerts and logs	Fields1 tool