Finance & Accounting · Finnhub + Claude

Finnhub MCP with Claude: Real-Time Market Data in Natural Language

The short answer

Claude works with Finnhub through the Model Context Protocol (MCP), an open standard that gives Claude secure, permissioned access to Finnhub's market data. Once connected with an API key, Claude can retrieve real-time quotes, company fundamentals, financial news and economic indicators in response to natural language questions, and your team controls exactly what it can read.

What is Finnhub?

Finnhub is a market data API that delivers real-time stock quotes, company fundamentals, financial news and economic data. Trading app developers, equity researchers, portfolio managers, fintech product teams and analytics groups use it to power dashboards, alerts and investment tools. It gives teams timely market intelligence without building and maintaining their own data feeds.

How Claude connects to Finnhub through MCP

The Model Context Protocol (MCP) is an open standard that lets Claude securely use other software, reading information and taking actions with permissions the organisation controls. Think of it as a universal adapter between Claude and the tool. Through a Finnhub MCP server or a direct API connection, Claude can reach Finnhub's data without anyone copying figures by hand.

Once connected, Claude can read live quotes, pull company financials, surface recent news for a ticker and query economic indicators. A portfolio analyst might ask for the latest earnings metrics for a set of companies and receive a summary in seconds. Claude answers from the data Finnhub returns, which keeps answers tied to source figures.

Permissions and approvals keep people in charge. Access is limited to the Finnhub endpoints your team chooses, the API key is stored securely and can be rotated or revoked at any time, and actions with consequences, such as sending an alert to a client or writing results into another system, can require a person to approve them first. Every request can be logged for audit.

What you can automate

  • Research Briefings for Analysts

    Analysts ask Claude for a briefing on a company covering its recent price movement, latest fundamentals and the most relevant news headlines. Claude gathers the figures from Finnhub and drafts a short note the analyst reviews before it goes into a research pack. This saves time spent switching between screens and copying numbers.

  • Market Movement Alerts

    Operations teams define watchlists and thresholds, and Claude checks Finnhub quotes against them. When a holding moves beyond a set level, Claude prepares a short alert with the relevant news attached and routes it to the right person. Staff decide what action to take, so the workflow supports judgement rather than replacing it.

  • Client Portfolio Summaries

    Relationship managers at wealth firms can ask Claude for a plain-English summary of how client holdings have performed and what has changed in the news. Claude draws current prices and company data from Finnhub and produces a draft update in the firm's house style. Advisers check the content before it is shared with a client.

  • Product Data for Fintech Apps

    Product teams building trading or investment apps use Claude to test how their data feeds respond to common questions and to generate sample content for in-app explanations. Claude can query Finnhub during development so teams spot gaps or formatting issues earlier. Engineers keep control of production data flows.

  • Economic Indicator Tracking

    Treasury and strategy teams ask Claude to pull economic indicators from Finnhub and place them alongside company news for a monthly review. Claude assembles the figures into a comparison table and flags notable changes. Analysts confirm the interpretation before the pack goes to leadership.

An end-to-end workflow: Weekly Holdings Review for an Investment Team

TriggerEvery Monday at 7:00, the team's scheduled task starts a review for the firm's model portfolio.

  1. Claude reads the list of holdings from the portfolio spreadsheet in Microsoft 365.

  2. Claude retrieves current quotes, key fundamentals and recent news for each holding from Finnhub.

  3. Claude flags holdings with moves beyond the agreed threshold and summarises the likely drivers from the news.

  4. Claude drafts a review document with a summary table and commentary, and saves it to the team's SharePoint folder.

  5. The portfolio manager reviews the draft, edits the commentary and approves it before it is circulated.

OutcomeThe team prepares its weekly review in a fraction of the usual time, with figures drawn directly from source data and a named person signing off every report. Typical teams spend far less time on data gathering and more time on analysis.

How we set it up

  1. Free Automation Audit

    We review your current research, reporting and client workflows to identify where Finnhub data would add the most value. You receive a clear list of priority workflows and a rough view of effort involved, at no cost.

  2. Secure API Key Connection

    We connect Claude to Finnhub using your API key, which is stored securely in a managed environment rather than in shared documents. We set up the connection through a Finnhub MCP server or a direct API connection, whichever suits your architecture, and limit access to the endpoints you need.

  3. Workflow Design and Approvals

    We design the end-to-end workflow across Finnhub and your other systems, such as Microsoft 365, your CRM or your ticketing tool. We define which actions run automatically and which need a person's approval.

  4. Testing Against Real Cases

    We test the workflow against past cases from your own team, comparing Claude's outputs with figures your analysts have already verified. We refine prompts, thresholds and formats until the results meet your standard.

  5. Launch, Monitoring and Training

    We launch with audit logs and monitoring in place, track usage and accuracy during the first weeks, and train your team on how to use and supervise the workflow. We adjust as your needs change.

Who benefits

Portfolio Managers
Faster weekly reviews with current figures and news pulled together in one draft for their approval.
Equity Research Analysts
Quicker company briefings that free time for deeper analysis and judgement.
Fintech Product Leads
A way to test and explain market data behaviour during development without slowing engineering teams.
Operations and Compliance Teams
Audit trails and approval steps that show who reviewed what and when.

Security and control

  • API keys are stored in a managed secrets store, never in chat logs or shared documents, and can be rotated or revoked at any time.
  • Access is limited to the Finnhub endpoints your workflow needs, following least-privilege principles.
  • Consequential actions, such as client communications or updates to other systems, require named human approval.
  • Every request and output is logged so your team can audit how figures were used.
  • Claude's answers are checked against source data during testing, and figures are always shown with their origin so staff can verify them.

For teams in the US, UK, India and the Middle East

Organisations in the United States and the United Kingdom can host the workflow in their chosen region to align with internal data policies and sector rules. In India, the Digital Personal Data Protection Act, 2023 shapes how personal data is handled, and we design workflows with that in mind. Firms in the United Arab Emirates and Saudi Arabia are subject to local data protection laws, and we select a hosting region that meets their requirements. Market data itself is generally public, but any client or personal data in the workflow is handled under the relevant regional rules.

Learn more about Claude and MCP

Frequently asked questions

What does the Finnhub MCP server do?

The Finnhub MCP server acts as a bridge that lets Claude request market data from Finnhub in a standard, controlled way. It can expose tools for retrieving real-time quotes, company financials, news and economic indicators. Your team decides which of these capabilities are available, so Claude only uses the data needed for the workflow.

Is connecting Finnhub to Claude secure?

Yes, when it is set up properly. Your API key is kept in a managed secrets store, access is limited to the endpoints you choose, and every request can be logged. Actions with consequences can require human approval. We review the setup with your IT team before launch.

What can Claude actually do with Finnhub?

Once connected, Claude can typically look up live quotes, retrieve company fundamentals, surface recent financial news and query economic indicators. It can then summarise, compare or format this information for your team. Claude does not act on the data alone unless you have designed and approved that step.

How long does setup take?

Timelines depend on how many workflows you want and how many systems they involve. A focused single workflow with one connected tool can often be live within a few weeks, including testing and training. We confirm the timeline during the free automation audit.

Which plan or permissions do I need for Finnhub?

Access depends on your Finnhub account and the data endpoints it includes, so you should check which access level covers the data your workflow needs. Connecting requires a valid API key. We help you confirm what your account supports during the audit, so you do not pay for data you will not use.

Book a free automation audit for your Finnhub workflows

Tell us which research, alert or reporting tasks take the most time, and we will show you how Claude and Finnhub could handle them. Book your free automation audit today, or write to our team to start the conversation. You can also e-mail contact@mindarora.com.