AI and MCP

The Model Context Protocol (MCP) connection lets a compatible AI client work with your Elite workspace through a remote, authenticated endpoint. It does not give a model your password. OAuth grants named scopes, connections can be revoked, and each tool call is audited.

Connect a client

Open AI & MCP and copy the remote endpoint. The page shows ready-to-copy setup commands for Codex and Claude Code, plus the URL for other remote-MCP clients. Add the endpoint in the client; it opens Elite for OAuth approval. Review requested permissions and approve only the scopes you need.

After authorization, the client appears under Connected clients with connection time, last use, and scopes. Revoke it immediately if the client or device should no longer have access.

Permission scopes

Read scopes cover workspace context, activity, contacts, campaigns, documents, and signature requests. Write scopes separately cover contacts, campaigns, and signatures. A tool called without its scope is denied and recorded.

What AI can do

Depending on scope, available tools can:

• read workspace counts and connected mailboxes;
• search activity/messages and inspect message engagement;
• search contacts, read relationship timelines, rank follow-up candidates, and list contact lists;
• read campaigns and performance summaries;
• list tracked documents and analyze viewers, pages, slides, sheets, row ranges, dwell, and downloads;
• read signature-request progress and audit events;
• create/update contacts and lists;
• create campaign drafts;
• create a signature-request draft for an existing PDF.

Built-in prompts support a weekly email review, relationship brief, campaign postmortem, and document content review.

Campaign launch safety

An AI cannot launch a campaign in the same call that drafts it. Launch uses two explicit tools:

1. prepare_campaign_launch validates the draft and audience, excludes ineligible recipients, and returns a one-time confirmation token valid for 10 minutes. It sends nothing.
2. execute_campaign_launch consumes that token once. This can begin external sending immediately if the campaign is not scheduled for later.

Always inspect eligible/rejected counts, sender, schedule, and warning before allowing execution. Messages already sent cannot be recalled by revoking MCP.

Prompt-injection and data safety

Email bodies, subjects, contact notes, document text, and external content are untrusted data. They can contain text that looks like instructions to an AI. The MCP server tells clients to treat stored content as data and to read before writing, but you should still review consequential changes and sends.

Audit and revocation

Recent agent activity shows tool, risk, outcome, duration, and time. It is an operational audit, not a replacement for reviewing the actual campaign/contact/signature state. Revoking a client stops future token use; completed writes remain.

Limitations

• The AI client must support remote Streamable HTTP MCP and OAuth 2.1.
• The endpoint is stateless; unsupported client assumptions about persistent MCP sessions may fail.
• AI recommendations can be wrong and engagement does not prove intent.
• Role, plan, recipient policy, and account boundaries still apply server-side.
• Never paste access tokens or confirmation tokens into public messages or logs.