Claude for Financial Services: What Anthropic's Free Plugin Does (and the Layer It Leaves to You)

TL;DR. Anthropic's free Claude for Financial Services plugins handle the analytical half of an advisor's work — financial plans, models, rebalancing, performance reports. They deliberately stop at the numbers. The half they leave to you is the compliant, client-facing words: the Reg BI rationale memo for the file, the Form ADV-consistent disclosure, the SEC Marketing Rule review of your LinkedIn post, the review-meeting follow-up in your voice. This guide maps the boundary and shows how to stack the two. Everything an AI drafts here is a draft — you and your compliance officer own what's sent.

In 2026 Anthropic released Claude for Financial Services — a set of free, open-source plugins (including a wealth-management pack) that install in Claude Cowork on any paid plan. They're genuinely good. If you advise clients, install them. But it's worth understanding precisely where they stop, because the place they stop is where most of an advisor's risk and relationship work actually lives.

What the free plugin does well

Anthropic's wealth-management plugin ships skills for the analytical core of the job:

• Client review prep — pulls performance and talking points before a meeting.
• Financial planning — retirement, education, and cash-flow projections.
• Portfolio rebalancing — drift analysis and tax-aware trade ideas.
• Client reporting — performance reports.
• Proposals — for prospective clients.
• Tax-loss harvesting — opportunity identification.

That's real leverage on the modeling and analysis. According to Orion's 2026 advisor survey, only about 6% of advisors are using agentic workflows at all — so even adopting these free plugins puts you ahead of most of the field.

Where it stops — and why it matters

The free plugin builds the plan and the numbers. It does not:

• Write the Reg BI rationale memo that documents why a recommendation was in the client's best interest, for the file.
• Draft a Form ADV-consistent disclosure in plain English for an email or proposal.
• Review your LinkedIn post, ad, or testimonial against the SEC Marketing Rule (Rule 206(4)-1).
• Turn a messy set of meeting notes into a books-and-records-ready memo.
• Write the market-volatility reassurance letter in your voice when a client is panicking.
• Run the prospecting, referral, and center-of-influence motion that grows the book.

This isn't a knock on the plugin — it's a scope choice. Analysis is the same for everyone; compliant client communication is specific to your registration, your firm's written supervisory procedures, your Form ADV, and your voice. That's the layer you (or a tool built for it) have to supply.

The mental model: analyst vs. practice

A useful way to think about it: the free plugin is the analyst. The practice — the words, the documentation, the growth — is the layer on top.

Anthropic's free plugin produces…	The layer it leaves to you…
A rebalance with tax-aware trades	The "here's why we're making this move" note to the client + the suitability documentation for the file
A financial plan	The plan-change explainer the client actually reads + a Reg BI rationale memo
A performance report	A reassurance letter when the report looks scary
A prospect proposal	The full prospecting → referral → COI motion around it

How to stack the two in practice

Here's a concrete weekly flow that uses the free plugin for analysis and a practitioner layer for everything client-facing:

1. Before a review, let the free plugin's client-review skill assemble the numbers, then turn that into a tight talking-points brief and agenda.
2. After the review, draft the follow-up email and a books-and-records meeting memo — in your firm's voice, not generic AI prose.
3. When you make a recommendation (a rollover, a product change), document the best-interest basis: capacity (Reg BI for a brokerage recommendation, the Advisers Act fiduciary standard for advice, or "which hat" if you're dually registered), the reasonably available alternatives you considered, the cost comparison, and the conflicts and how you addressed them.
4. Before anything goes public — a post, an ad, a seminar invite — run it against the Marketing Rule: testimonials and endorsements need clear-and-prominent disclosures; performance needs net figures and the prescribed 1-, 5-, and 10-year periods; "best advisor" superlatives are red flags.
5. Grow the book with referral asks, center-of-influence outreach, and educational content — all written to stay inside the same compliance lines.

The AI Career Lab's Advisor tools and the Financial Advisor AI Prompts vault are built to be exactly this practitioner layer — designed to stack on top of Anthropic's free plugin rather than duplicate it.

The non-negotiable: a human (and your CCO) owns the output

One rule governs all of this. Whatever an AI drafts — free plugin or practitioner layer — is a draft. It is not legal, tax, investment, or compliance advice, and it does not make anything compliant. Principal approval of communications (FINRA Rule 2210) and your firm's best-interest and supervisory obligations (Reg BI) remain yours. Every output should be reviewed and approved by your compliance officer or supervisory principal before it's used in a regulatory filing or sent to a client. The tool collapses the blank-page time; it never collapses the accountability.

Used that way — analysis from the free plugin, compliant words from a layer built for it, and a human owning every decision — AI stops being a novelty in a browser tab and starts being the most reliable associate in your practice.

---

Anthropic, Claude, Cowork, and Claude for Financial Services are marks of Anthropic, PBC, referenced here only to describe compatibility. The AI Career Lab is not affiliated with or endorsed by Anthropic. This article is educational and is not legal, tax, investment, or compliance advice.