Claude's Finance Agents: Your Free Wall Street Analyst 🛠️

Anthropic just launched 10 AI finance agents. Three of them give everyday investors the same tools Wall Street analysts charge thousands for: a Market Researcher that monitors your entire portfolio, a Model Builder that creates full stock valuations in Excel, and an Earnings Reviewer that breaks down every earnings call.

This guide is for educational purposes only. Nothing here is financial advice. Always do your own research and consult a qualified financial advisor before making investment decisions.

On May 5, 2026, Anthropic released 10 finance agent templates for Claude. These are not basic chatbot prompts. They are full agent systems with data connectors, sub-agents, and domain-specific skills that handle real financial workflows.

The agents connect to professional data sources like FactSet, S&P Global, Moody's, MSCI, and the London Stock Exchange Group. The same data Wall Street firms pay six figures a year to access. Claude can now pull from these sources, analyze the data, and deliver institutional-quality research to anyone.

Research agents

• Market Researcher - monitors sectors, synthesizes news, filings, and broker research
• Model Builder - creates and maintains financial models from filings and data
• Earnings Reviewer - analyzes earnings transcripts, updates models, flags thesis changes
• Pitch Builder - creates target lists, comparable analysis, drafts pitchbooks
• Meeting Preparer - builds client and counterparty briefs before calls

Operations agents

• Valuation Reviewer - reviews valuations against comparables and methodology
• GL Reconciler - reconciles general ledger accounts, performs NAV calculations
• Month-End Closer - manages month-end close workflows and task sequencing
• Statement Auditor - reviews financial statements for consistency and audit readiness
• KYC Screener - gathers entity files, reviews docs, prepares compliance escalations

Most of these are built for finance professionals. But three of them are incredibly useful for everyday investors who want to do their own research without paying for a Bloomberg terminal or a financial advisor. Those are the ones we are going to set up.

There are two parts to this setup. First you install the Finance plugin, which gives you all 10 finance skills. Then you add the three agent marketplaces from GitHub, which gives you the full Market Researcher, Model Builder, and Earnings Reviewer agents. Takes about five minutes.

1. Go to Customize. Open Claude (either claude.com or the Cowork desktop app). On the left sidebar, click Customize. You need a paid plan (Pro, Max, Team, or Enterprise) for plugins.
2. Install the Finance plugin. Under Personal Plugins, hit the + button. Then click Browse Plugins. Search for Finance and add it. This installs all 10 finance skills.
3. Add the agent marketplaces from GitHub. Back in Personal Plugins, hit + again, click Create Plugin, then Add Marketplace. Select Add from a repository and paste in the GitHub link for each agent.

GITHUB REPOSITORY LINKS

github.com/anthropics/financial-services/market-researcher

github.com/anthropics/financial-services/model-builder

github.com/anthropics/financial-services/earnings-reviewer

The finance agents are fully open source at github.com/anthropics/financial-services. You can inspect how they work, customize them, or deploy them through the API to build your own tools.

Works in three places: Claude.ai (web), Cowork (desktop), and Claude Code (command line). Same agents, same capabilities, pick whichever you prefer.

Your personal analyst covering every stock you own.

This is the agent I use the most. You paste in the stocks you hold and Claude pulls every news story, every analyst rating, every SEC filing, every major announcement, all automatically. It is like having a research analyst who covers your entire portfolio and never sleeps.

What it gives you:

• News monitoring - every relevant story, filtered for what actually matters
• Analyst ratings - upgrades, downgrades, price target changes, and reasoning
• SEC filings - 10-K, 10-Q, 8-K filings parsed and summarized in plain English
• Sector trends - industry-wide developments that affect your holdings
• Competitive intel - what competitors are doing and how it impacts your stocks
• Risk alerts - regulatory changes, lawsuits, management departures, red flags

Prompts to try

PORTFOLIO BRIEFING

I own shares in AAPL, NVDA, MSFT, GOOGL, and AMZN.

Give me a full briefing on all five stocks. What happened this week? Any analyst rating changes? Any news I should know about? Flag anything that changes the investment thesis for any of these positions.

DEEP DIVE ON ONE STOCK

Give me a complete research report on Tesla (TSLA). Cover: recent earnings, analyst consensus, key risks, competitive position, upcoming catalysts, and whether the current valuation makes sense based on growth.

SECTOR SCAN

I am interested in investing in the AI chip sector. Give me an overview of the key players, market share, revenue growth, and which companies are best positioned for the next 2-3 years. Include NVDA, AMD, INTC, AVGO, and any others I should be watching.

WEEKLY WATCHLIST

Here is my watchlist: PLTR, SNOW, CRWD, NET, DDOG.

Rank them by investment attractiveness right now. For each one, give me: current price context, recent developments, analyst sentiment, and the one thing I should be watching most closely.

Tip: ask Claude to give you a weekly briefing every Monday morning. You can set this up as a recurring task. Paste your portfolio once and Claude will monitor it for you.

Full financial models that analysts charge thousands to create.

This is the one that blew my mind. You give Claude any stock and it builds you a complete financial model in Excel. Revenue projections, expense forecasts, DCF valuation, sensitivity analysis, the whole thing. Investment banks charge clients thousands of dollars for models like this.

What you get: full revenue and expense projections, DCF valuation, comparable analysis (P/E, EV/EBITDA, P/S), sensitivity tables across different growth and margin scenarios, and a clean assumptions page so you can adjust scenarios yourself.

Prompts to try

BUILD A FULL MODEL

Build me a complete financial model for Microsoft (MSFT) in Excel. Include 5 years of revenue and expense projections, a DCF valuation, a comparable analysis vs peers, and a sensitivity table. Show me what the stock is worth at different growth and margin assumptions.

COMPARE TWO STOCKS

I am deciding between CrowdStrike (CRWD) and Palo Alto Networks (PANW). Build me a side-by-side model comparing revenue growth, margins, valuation multiples, and implied upside for both. Which one is cheaper relative to its growth?

SCENARIO ANALYSIS

I own Amazon (AMZN). Build me a model that shows what happens to the stock price if: (a) AWS growth slows to 15%, (b) AWS growth stays at 25%, (c) AWS growth accelerates to 35%. Hold everything else constant.

Important: financial models are only as good as their assumptions. Claude builds the model, but you should always review the assumptions and adjust them based on your own research. Do not blindly trust any model output, whether it comes from AI or a human analyst.

Know exactly what management said and whether your thesis still holds.

Earnings calls are where companies tell you what is actually happening. But they are long, full of corporate language, and most investors do not have time to listen to 20 calls a quarter. The Earnings Reviewer reads the full transcript and tells you exactly what matters.

Prompts to try

EARNINGS RECAP

Apple (AAPL) just reported earnings. Review the full transcript. Tell me: what beat and missed expectations, what changed in their guidance, and should I be concerned about anything?

THESIS CHECK

I own Nvidia (NVDA) because I believe AI infrastructure spending will continue to grow for at least 3 more years. Review their latest earnings call and tell me: does management's commentary support or undermine this thesis? Be specific.

COMPARE EARNINGS ACROSS QUARTERS

Pull the last 4 earnings calls for Meta (META). How has management's tone changed over the past year? Are they getting more or less confident about their AI investments? Is ad revenue re-accelerating?

EARNINGS SEASON BATCH

This week, AAPL, AMZN, GOOGL, and META all reported earnings. Give me a summary of each one. Which beat expectations? Which missed? Which had the most concerning guidance? Rank them from most bullish to most bearish based on earnings quality.

Tip: the Earnings Reviewer works best when you tell it your thesis first. Instead of just saying "review this earnings call," say "I own this stock because of X. Does this earnings call support or undermine that?" The analysis gets much more targeted and useful.

Instantly see whether a stock is cheap or expensive versus its real peers.

This is the /comps skill inside the finance plugin. Give it a company and it pulls the actual competitors, then builds a full benchmarking table: valuation multiples (P/E, EV/EBITDA, P/S) and operating metrics (revenue growth, margins) side by side. You see at a glance whether the stock is trading cheap or rich against the companies it truly competes with.

Prompts to try

PROMPT

Run a comparable company analysis on Nvidia (NVDA). Pull its real peers and build a table of valuation multiples (P/E, EV/EBITDA, P/S) and operating metrics (revenue growth, gross and operating margin). Tell me whether NVDA looks cheap or expensive versus the group, and why.

PROMPT

Compare Coca-Cola (KO) and PepsiCo (PEP) on valuation and growth. Which one is cheaper relative to what it's actually delivering?

PROMPT

Build me a comps table for the 6 largest US airlines and flag the one that looks most mispriced, with your reasoning.

Tip: a stock can look expensive on its own and cheap next to its peers, or the reverse. Comps is how you tell the difference. Ask Claude which peers it chose and whether they're genuinely comparable.

Read any company's financials and surface the risks before you invest.

The finance plugin's analysis and audit skills let Claude go through a company's actual statements, income statement, balance sheet, cash flow, and flag what matters: rising debt, thin cash, hidden or growing liabilities, one-time items propping up earnings, and numbers that don't line up. It's the check you'd want a careful analyst to run before you put money in.

Prompts to try

PROMPT

Go through [COMPANY]'s latest financial statements. Flag the biggest risks, any hidden or growing liabilities, anything unusual in the cash flow, and any numbers that rely on one-time items or don't add up. Tell me what would make you cautious.

PROMPT

Look at [COMPANY]'s balance sheet and debt load. Is the debt a problem? How does it compare to their cash flow and to their peers?

PROMPT

Read [COMPANY]'s last 10-K. What are the three risks buried in the filing that a headline reader would miss?

Tip: pair this with the Earnings Reviewer. The financials show you the shape of the business; the earnings call shows you what management is saying about it. Together they catch a lot.

Turn your own spending into a budget, a forecast, and a list of leaks.

Quick note: this one isn't part of the finance-agents plugin, that's built for investing. This is just Claude working with your own numbers, and it needs no setup. Export your recent transactions from your bank or budgeting app (a CSV, or even a screenshot), hand them to Claude, and it builds a real budget, forecasts your cash flow for the months ahead, and points out exactly where your money is quietly leaking every month.

Prompts to try

PROMPT

Here are my last 3 months of transactions [paste or upload]. Build me a monthly budget by category, show my average income and spending, and tell me where my money is leaking, the recurring charges and categories I'm overspending on.

PROMPT

Based on this spending, forecast my cash flow for the next 6 months. If nothing changes, where do I end up, and what one or two changes would move the needle most?

PROMPT

Find every subscription and recurring payment in these transactions and total them up. Which ones am I probably not using?

Tip: don't paste full card numbers, and if you're privacy-conscious, remove or rename anything sensitive first. Claude only needs the category totals, not your account details.

The real value comes when you chain these agents together. Here are some workflows I use regularly that combine multiple agents in a single conversation.

COMPLETE DUE DILIGENCE

I am considering buying shares of Palantir (PLTR). Run a full due diligence:

1. Market research: news, analyst ratings, competitive position, key risks
2. Build a full financial model with DCF valuation and sensitivity table
3. Review the most recent earnings call and tell me if the thesis holds
4. Give me a final recommendation: buy, hold, or pass, with reasoning

EARNINGS PREP

NVDA reports earnings next week. Help me prepare:

1. What are consensus estimates for revenue and EPS?
2. What are the three things to watch most closely?
3. Build me a quick model showing what the stock is worth at different revenue growth scenarios.
4. What did management guide for this quarter on the last earnings call?

STOCK SCREENING

Find me 5 stocks I am probably not looking at that meet these criteria: market cap $5B-$50B, revenue growth above 25%, improving margins, and a strong competitive moat. For each one, give me a quick overview of why it is interesting and what the risks are. Avoid the obvious mega-caps.

The three agents above are the ones most useful for individual investors. The finance plugin includes 7 more agents designed for finance professionals.

• Pitch Builder - creates target lists, comparable analysis, and drafts pitchbooks for bankers
• Meeting Preparer - builds client and counterparty briefs before calls
• Valuation Reviewer - reviews valuations against comparables and methodology
• GL Reconciler - reconciles general ledger accounts and performs NAV calculations
• Month-End Closer - manages month-end close workflows and task sequencing
• Statement Auditor - reviews financial statements for internal consistency, GAAP compliance, and audit readiness
• KYC Screener - handles Know Your Customer checks, gathers entity documentation, reviews it against regulatory requirements, and prepares escalation reports

Part of what makes these agents so powerful is the data they can access. Anthropic partnered with FactSet, S&P Global, Moody's, MSCI, the London Stock Exchange Group, and expert networks for primary research and sector intelligence.

Free vs premium data: many of these data providers require their own subscriptions. The agents work with publicly available data by default (SEC filings, news, free financial APIs). If you have a FactSet or Bloomberg subscription, you can connect it for deeper data. Most individual investors will get plenty of value from the free data alone.

1. Always state your investment thesis. "Analyze NVDA" is fine. "I own NVDA because I believe AI infrastructure spending will grow 30%+ for the next 3 years. Analyze whether this thesis still holds." is ten times better.
2. Ask for the bear case. After any bullish analysis, ask Claude to argue the opposite side. You will catch risks you would have missed.
3. Chain agents in one conversation. Research → model → earnings review → recommendation. Each agent builds on the last.
4. Re-run analyses periodically. Markets change. A thesis that held last quarter may not hold this quarter. Use the same prompts every few weeks.

These agents are research tools, not financial advisors. They can analyze data, build models, and summarize earnings calls, but they cannot predict the future. Never invest based solely on AI output. Always do your own due diligence.

Wall Street tools. Zero Wall Street fees.