Morningstar MCP Server
Connect trusted data to your AI workflows with the Morningstar MCP Server.
About
General-purpose AI tools answer investment questions from training data that can be stale, unsourced, or simply wrong, which makes their output hard to trust or audit. The Morningstar Plugin serves investment teams, research leads, and platform owners who are scaling AI across research, portfolio review, and reporting, by grounding answers in Morningstar data and analyst research that can be traced and checked.
Common Use Cases
- Research and shortlist investments Screen funds and stocks against your own criteria, then compare the candidates using Morningstar analyst research and data.
- Review a client portfolio Analyze a portfolio's asset allocation, sector and style exposure, and overlapping holdings, and summarize what changed and why.
- Draft reports and proposals Produce client-ready reports and proposals grounded in Morningstar performance, holdings, risk metrics, and analyst views.
- Monitor concentration risk Look through funds to their underlying holdings to catch overlaps and concentrated positions, and find alternatives when risks appear.
Capabilities
- Investment screening Find ETFs, open-end funds, closed-end funds, collective investment trusts, and stocks that match criteria such as fees, returns, ratings, and valuation. Criteria combine with AND logic only, and screens work best on numeric data points.
- Analyst research The latest Morningstar analyst reports, notes, and quantitative reports for stocks, funds, and ETFs, including fair value estimates, economic moat, and Medalist Rating pillar analysis. Only the most recent research is available, typically within 24 hours of publication.
- Investment data Values for more than 1,300 data points (as of July 2026) across funds and stocks, with history for select data points. How much history is available varies by data point.
- Fund holdings The top 10 to 100 holdings of open-end funds, closed-end funds, and ETFs, with portfolio weights, for several funds in one request. Only the largest holdings are returned, not the full portfolio.
- Portfolio analysis Morningstar X-Ray analysis of a custom portfolio or a single fund, covering asset allocation, sectors, regions, style, fixed income, and stock overlap. Comparisons use the S&P 500 as a fixed benchmark.
- Market commentary Answers drawn from Morningstar's editorial content, including market commentary, investment themes, and explanations of Morningstar methodology. Articles from 2022 onward only.
- Research skills Five ready-made workflows for fund screening, fund comparison, fund summaries, Medalist Rating analysis, and finding data points, with consistent output formatting and guardrails against inferred data. Four of the five focus on funds.
For every tool, its limits, and example questions, see Tools.
Supported Platforms
The plugin and the Morningstar MCP Server are available on different platforms.
- Morningstar Plugin ChatGPT, through the marketplace. Claude, installed manually, because the plugin isn't yet listed in the Claude plugin marketplace.
- Morningstar MCP Server Claude, ChatGPT, Microsoft Copilot Studio, and Microsoft Foundry.
- Custom agents Microsoft Copilot Studio, through the official Morningstar connector.
Coverage
- Universe More than 200,000 global securities TK: as-of date, and more than 1,300 data points as of July 2026.
- Investment Types ETFs, open-end funds, closed-end funds, collective investment trusts, and stocks. Analyst research covers stocks, open-end funds, and ETFs.
- Geographies Global securities. Stock research has global coverage, and fund research covers more than 20 domiciles, including the US, Canada, Japan, Australia, France, the UK, and Luxembourg.
- History Investment data: available for select data points, varying by data point. Analyst research: latest reports and notes only, with no archive. Market commentary: articles from 2022 onward.
- Update Frequency Analyst research: typically within 24 hours of publication.
- Entitlements Access follows your Morningstar licence: securities and data points outside it aren't available through the server.
Results depend on your AI client. Answers are only as reliable as the client's use of the tools. The client decides which tools to call and how to combine the results, so the same question can produce different answers on different models. Data point lookups by name are also limited to a mapped subset of the 1,300+ data points.
Limits
- Results depend on the AI model Morningstar recommends Claude models, or OpenAI GPT-5.1 or later, for reliable tool calling. Models with weaker tool calling may skip steps or misread results, so test your chosen model against your own workflows before rolling it out.
- Names must resolve to Morningstar IDs Every request starts by matching names and tickers to Morningstar identifiers, and many securities and data points have similar names. Ambiguous requests can resolve to the wrong one, so prompts and agent instructions should use tickers where possible.
- Mapped data points only Screening and data retrieval by name are limited to the data points the lookup tool has mapped. A data point outside that subset can't be requested by name. TK: whether it can be requested by ID.
- Screening logic Criteria combine with AND only, using equals, greater than, and less than. Questions that need OR conditions require separate screens whose results are then combined.
- Latest research only Analyst research returns the most recent report and notes, so questions about how Morningstar's view has changed over time can't be answered from research alone.
- Fixed benchmark Portfolio analysis compares against the S&P 500 only, so non-US and multi-asset portfolios are measured against a US equity benchmark.
- Top holdings only The holdings tool returns up to 100 holdings per fund, so look-through analysis built on it covers the largest positions rather than the full portfolio.
- Usage limits Rate limits are determined by your contract.
Access
- Licence Required Yes. Securities and data points are available only if they're part of your Morningstar licence. Morningstar account manager can confirm what's included.
- Self-Serve Trial Access See Trial Access for more information.
- Authentication Sign in through OAuth 2.0 with your Morningstar credentials. Your AI client prompts you the first time it connects to the server.
Looking for Something Else?
- Data access via APIs APIs