Lumonic, an AI-powered portfolio monitoring platform for private credit and private equity firms and a PitchBook company, today announced the general availability of its Model Context Protocol (MCP) server for all customers. The server connects Claude, ChatGPT, and any MCP-enabled assistant a firm has approved to its Lumonic workspace. This connection enables users to query Lumonic portfolio data directly from their preferred workspace, including which companies have not submitted financials or which borrowers are approaching a covenant breach. With this launch, Lumonic is also releasing its own MCP Library, which includes tested prompts and downloadable artifacts for private credit, private equity, and venture capital workflows, available at lumonic.com/mcp/library.
MCP is an open standard that allows AI assistants to read data from systems an organization already uses. The Lumonic MCP server provides read-only access to periodic financials, covenant status, reporting status, and tracked KPIs, mapped to each user’s existing Lumonic permissions. Every figure the server returns includes a link to its source document, down to the specific cell or location on the page. Lumonic’s underlying data extraction is reviewed by each firm’s own team before it is made available to an AI assistant.
With the Lumonic MCP server, finance teams can use an AI assistant to:
- Identify outstanding items: View which companies have not submitted financials, along with what is missing and how long it has been outstanding
- Monitor covenant compliance: Compare each borrower’s tested threshold to its actual figure to flag covenants at risk of breach
- Generate reporting materials: Produce a one-pager for each portfolio company ahead of quarterly review
- Analyze portfolio exposure: Pull exposure data across the portfolio by sector, vintage, or other tracked dimensions
- Maintain live, auditable workbooks: Write Lumonic formulas directly into Excel so figures refresh automatically and remain traceable to source documents
The suite of features and capabilities is designed to reduce the manual work involved in portfolio reporting while preserving the source-level audit trail required for institutional use.
“Any AI assistant can produce an EBITDA figure in seconds. Proving it is the hard part,” said Kevin Hsu, CEO of Lumonic. “Every number the Lumonic MCP returns links back to its source document, down to the exact cell, so the answer holds up in front of an auditor or an investment committee. From day one, Lumonic was built with an AI-first mindset. Portfolio monitoring in private credit and private equity is an incredibly manual process, and Lumonic has already been streamlining it for our customers for years. This release makes that even easier.”
Together with PitchBook’s Premium Connector, clients of both PitchBook and Lumonic can easily access private market data alongside their own portfolio data, all from a single workspace or AI prompt. For example, the first MCP Library artifact built on both servers is a valuation workbook that pulls PitchBook comps and a firm’s own portfolio financials into the same auditable Excel model. Lumonic was acquired by PitchBook, a Morningstar company, in 2025. The acquisition empowers private market investment professionals to manage and analyze their portfolios alongside PitchBook’s industry-leading private capital data. Since the acquisition, Lumonic has more than doubled its customer base year over year, serving firms across private credit, private equity, and venture capital. The MCP server has been in production since early this year, with firms using it daily.
The MCP server is included for all Lumonic customers. The quickstart, prompt library, and security documentation are available at: lumonic.com/mcp
