💹 Claude Finance Agents — 3 Tools That Collapse the Retail vs Institutional Gap

/earnings-analysis     → 8-12 page earnings update from any transcript
/initiating-coverage   → Full initiation report via 5-task workflow
/earnings              → Shorter, faster quarterly earnings analysis
/initiate              → Initiating coverage entry point
/screen                → Stock screen or investment idea generator
/sector                → Sector overview report
/sector-overview       → Full industry overview write-up
/thesis                → Create or update an investment thesis
/thesis-tracker        → Track performance of existing theses
/catalysts             → View or update the catalyst calendar
/catalyst-calendar     → Track upcoming catalysts on your watchlist
/earnings-preview      → Pre-earnings setup — what to watch before the print
/idea-generation       → Surfaces investment ideas in a sector or theme
/model-update          → Pushes new numbers into your tracking model
/morning-note          → Drafts a morning note for your records
	```</code></pre></div></details>
💬 How to Use It
AAPL just reported. Drop the transcript into /earnings-analysis.
Get back a ten-page institutional research note covering:
Revenue mix
Segment performance
Guidance changes
Services growth trajectory
Every risk management buried in the call
Same format a Goldman analyst would write. Done in three minutes.
	```</code></pre></div></details>
⚠️ Honest Limit
Garbage in, garbage out. The note is only as good as what you feed it — a transcript, a 10-K, a press release. Without inputs, it sits idle. It won't tell you whether to buy. It'll tell you what changed and where the risks sit. The decision is still yours.

🔍 02 — Earnings Reviewer
Tier 1 · Game-changer · Agent · 6 commands
If Equity Research is the workspace, this is the scalpel. Six commands, one mission: a stock you own just reported — read the call, update the model, draft your note. End-to-end on a single quarterly print.
📋 All 6 Commands
/earnings-analysis   → Full earnings update report from a transcript
/earnings-preview    → Pre-earnings setup — what consensus expects, what to watch
/model-update        → Pushes new revenue, margin, and guidance into your spreadsheet
/morning-note        → Short morning note summarising the quarter
/audit-xls           → Audits a spreadsheet for formula errors and mistakes
/xlsx-author         → Outputs a clean .xlsx file
	```</code></pre></div></details>
💬 Copy-Paste Prompt
You are a buy-side analyst. I'm going to give you an earnings transcript.
Your job:
Summarise what management said about revenue, margins, guidance, and key risks — in plain English.
Flag any language that is vague, overly optimistic, or contradicts prior guidance.
Tell me whether a long-term holder should feel more or less confident after this quarter, and why.
Here is the transcript:
[PASTE TRANSCRIPT HERE]
	```</code></pre></div></details>
⚠️ Honest Limit
Narrower than Equity Research — no sector views, no screening, no thesis tracking. If you own a handful of stocks and just want to stay on top of their earnings, this is the cleanest tool. If you want broader workflows too, install Equity Research instead — or both.

🧮 03 — Financial Analysis
Tier 1 · Game-changer · Skill bundle · 18 commands
The math layer. DCFs, comps, three-statement models — everything you need once you've decided a company is worth modelling.
The clever part: the audit and debug commands. Paste in any model you built or downloaded and get every formula error, broken link, and inconsistent assumption flagged in seconds. That alone justifies the install.
📋 7 Retail-Relevant Commands
/dcf                → DCF valuation with comps-informed terminal multiples
/comps              → Comparable company analysis with trading multiples
/3-statement-model  → Income statement, balance sheet, cash flow from filings
/comps-analysis     → Deeper institutional-grade comps with operating metrics
/audit-xls          → Audits any spreadsheet for formula accuracy and errors
/debug-model        → Debugs a financial model for broken assumptions
/deck-refresh       → Updates a presentation with new numbers
	```</code></pre></div></details>
💬 Copy-Paste Prompt
Build me a DCF valuation for [COMPANY / TICKER].
Include:
A base, bull, and bear case for revenue over 3–5 years
Key assumptions for each case (growth rate, margins, exit multiple)
An estimated fair value range per share
The top 3 risks that could break the bull case
Format it as a clean table I can copy into Excel
Use publicly available financials and be transparent about every assumption.
	```</code></pre></div></details>
⚠️ Honest Limit
Steepest learning curve of the four. DCF and three-statement modelling assume you understand what they are — but worth using even if you don't. Claude explains its assumptions as it goes, and you'll learn faster building one DCF than reading ten articles about them. Start with /audit-xls if you're new.

🌐 04 — Market Researcher
Tier 1 · Game-changer · Agent · 5 commands
The wide-angle lens. Where Earnings Reviewer is a microscope on one company, this zooms out across an entire sector or theme. Use it when you don't know what you're investing in yet — when you've heard a theme is hot and want to understand the landscape before picking a stock.
📋 All 5 Commands
/sector-overview        → Full industry overview — players, dynamics, where money flows
/competitive-analysis   → Maps the competitive landscape inside a sector
/comps-analysis         → Peer comps across companies in the same sector
/idea-generation        → Surfaces 5-10 investment ideas inside a theme
/pptx-author            → Outputs sector deck to PowerPoint
	```</code></pre></div></details>
💬 Copy-Paste Prompt
I want to understand [SECTOR / THEME] as an investment opportunity.
Give me:
A sector overview — key players, market dynamics, where money is flowing
The 3-5 most important trends driving growth or risk right now
5 specific stock ideas worth doing deeper research on
Which 2-3 names look most attractive on valuation vs peers right now
Keep it scannable. I&#x27;ll use this as a starting point for deeper work.
	```</code></pre></div></details>
⚠️ Honest Limit
Only five commands — light by design. Once you've narrowed to a stock, switch to Earnings Reviewer or Equity Research for the deep work. Think of Market Researcher as the funnel that feeds the other two, not a full research tool on its own.

🚫 What These Agents Won't Do
Read this section twice. It's the part that protects you from the tools.
01 — They won't tell you what to buy
Run /dcf on TSLA and you'll get a fair-value range, not a recommendation. Run /earnings-analysis on META and you'll get a complete read of the quarter, not a position to take. The agents surface the inputs to a decision. Pulling the trigger is still your job — and it should be.
02 — Their data has a cutoff
Most plugins lean on Claude's knowledge plus whatever you paste in. If you ask about a stock without giving it a current transcript or filing, you may be working with months-old context. Rule of thumb: if the analysis cites a specific number, the analysis is only as fresh as the document you fed it. Always paste the actual filing or transcript.
03 — Garbage in, garbage out
A DCF is only as good as the assumptions you put in. If you use a 15% perpetual growth rate, you'll get a clean, professional, completely useless valuation. The skill that matters with these tools isn't prompting — it's knowing when an output looks plausible but is built on a bad input.
04 — They don't replace judgement
These agents reduce the time from question to first answer by 90%. They don't reduce the time you should spend on the decision. The temptation with any tool this fast is to use the speed to do more work. The right move: use the speed to do better work on fewer names. Read more, decide less, hold longer.

🔗 How to Install
Platform: claude.ai → Integrations / Marketplace
Recommended installs: Equity Research + Earnings Reviewer to start
Add Financial Analysis once you're comfortable modelling
Add Market Researcher for sector exploration and idea generation