Affinity MCP for Private Capital: A Deal Team's Guide

The Affinity MCP (Model Context Protocol) server allows private capital deal teams to connect AI assistants like Claude directly to their firm's live relationship intelligence data. This enables you to ask natural language questions to find the strongest introduction paths, get pre-meeting briefings, and summarize relationship histories, streamlining deal sourcing and management workflows.

The challenge: Finding relationship context buried in your CRM

For a private capital firm, the network is the most valuable asset. Affinity is often the system of record for this network, tracking every interaction, connection, and relationship score. However, extracting actionable intelligence from it can be a manual, time-consuming process. Associates and partners often spend hours clicking through profiles, reading past notes, and cross-referencing lists to piece together the answer to a simple question: "Who is the best person to introduce us?"

Preparing for a meeting involves a similar trawl through contact records, email threads, and meeting notes to build a complete picture. This manual work is inefficient and risks missing critical context that could make or break a conversation.

Illustrative example: Using Affinity MCP for a pre-call briefing

A private equity associate is preparing their Managing Director for a call with the CEO of a potential acquisition target. Instead of spending 30 minutes manually searching the firm's Affinity instance, the associate uses their AI assistant.

Prompt to AI Assistant: `"Generate a pre-meeting brief for my call with Sarah Jenkins, CEO of Innovate Inc. What is our firm's complete relationship history with her and her company? Who knows her best, what was our last interaction, and are there any open follow-ups?"`

The AI assistant, using the Affinity MCP server, securely queries the firm's live CRM data. It doesn't use a stale data export or a public web search. It accesses the firm's single source of truth in real time and returns a concise, sourced summary.

AI-Generated Brief:

• Strongest Relationship: Your partner, David Chen, has the strongest connection (Affinity relationship score: 95). They have exchanged 28 emails and have 14 mutual connections.
• Last Interaction: David last met with Sarah three months ago at the Global Tech Summit. His notes indicate they discussed go-to-market strategies for Europe.
• Company-Wide History: Our firm has interacted with six other executives at Innovate Inc. over the last 18 months, primarily through your associate, Maria Rodriguez, in the context of the '2025 SaaS Leaders' list.
• Open Items: There is a note logged for David to follow up with Sarah regarding an introduction to their head of product.

With this information instantly available, the associate can prepare their MD in minutes. The source note for the Affinity MCP mentions supported "writes," so the associate could potentially follow up with: `"Log a new note on the Innovate Inc. record that we have a follow-up call scheduled for next Tuesday at 10 AM."` This updates the CRM instantly, ensuring the data is always current.

Pre-call briefing checklist using Affinity MCP

Use this checklist to ensure your team is fully prepared for any external meeting by querying your Affinity data via an MCP-enabled AI assistant.

• Verify the contact's current title and company.
• Identify the strongest relationship owner within your firm and their Affinity relationship score.
• Review the date, participants, and summary of the last interaction with the contact and their organization.
• List all team members who have had contact with anyone at the target company.
• Summarize key themes, questions, and outcomes from past meeting notes.
• Check for any open tasks or promised follow-ups associated with the relationship.
• Ask for a summary of recent notes or logged emails to understand current context.
• (Post-call) Use the AI assistant to draft a meeting summary and log it as a new note in Affinity.

Prerequisites and limitations

To use an AI assistant with your firm's Affinity data, several components must be in place.

Prerequisites:

• An active Affinity subscription (Growth, Scale, or Enterprise tiers).
• An AI assistant that supports MCP, such as Claude.
• The Affinity MCP server must be enabled for your account. This is provided by Affinity, not a third party.
• Users must authenticate with their own Affinity credentials, ensuring they can only access data permitted by their existing user rights.

Limitations:

• Access, Not Ownership: MCP provides a new way for your team to access its own CRM data. It does not establish ownership of the data or grant you the right to sell or license it. Data licensing is a distinct legal and commercial process that requires explicit company authorization. See our guide on MCP and data licensing rights.
• Data Quality: The AI's responses are only as good as the data in your Affinity instance. Incomplete or outdated records will lead to incomplete or outdated answers.
• Scoped Actions: Write capabilities are typically constrained to specific, safe actions like adding a note or creating a task. MCP is not a tool for bulk data modification or system administration. Always check the current vendor documentation for a precise list of supported actions.
• No Permission Escalation: MCP respects Affinity's underlying permissions model. A user cannot ask the AI assistant to retrieve information they would not be able to see by logging into Affinity directly.

Questions to ask your software provider or implementation team

When evaluating or implementing an Affinity MCP integration, ask these questions:

1. Which specific Affinity data objects can be queried? (e.g., People, Organizations, Lists, Notes, Opportunities, Email/Calendar data)
2. What are the precise "write" actions supported? Can we create, update, or only append information like notes?
3. How does the MCP server handle our firm's custom fields and configured lists?
4. What authentication method is used to connect a user's AI session to their Affinity account? Does it support our firm's SSO provider?
5. What kind of audit logs are generated for MCP queries, and how can we access them?
6. Can we define which lists or data types are exposed to the AI assistant via the MCP server?

Next step with SourceX

Using MCP with Affinity is a powerful way to unlock the internal value of your firm’s proprietary network data for sourcing and diligence. A separate, and often larger, opportunity exists in the external value of the operational data held by your portfolio companies.

AI labs and data buyers are actively seeking high-quality business datasets to train next-generation models. If your firm invests in or advises US-based operating companies with 50-500 employees, their anonymized operational data—from ERP, CRM, and other systems—may be a candidate for licensing.

Use the Portfolio Data Opportunity Scanner to run a permissioned, high-level screen of your portfolio companies to identify potential fits for data licensing. For each introduced company that completes a data transaction, your firm receives 25% of the platform fees SourceX collects, up to $100,000 per referred company. This reward is a share of SourceX's fee and is entirely separate from the licensing proceeds paid to the portfolio company itself.

Related MCP guides

• MCP for Private Equity: Unlocking Portfolio Data and Accelerating Deal Workflows
• MCP for Warm Introductions: Find the Best Path in Your Firm's CRM
• Chronograph MCP: Querying Private Capital Portfolio Data in Claude
• All MCP resources

Sources

• Anthropic finance agents (May 5 2026)
• Chronograph MCP launch (October 28 2025)
• Affinity private-capital MCP (Updated July 16 2026)

Vendor capabilities change. Check current official documentation before relying on any product detail.