What does this skill do, and when should you use it?
Client Review Prep is designed for wealth-management advisors preparing quarterly reviews, annual checkups, or ad hoc client meetings. It consolidates client and household details, account types, total AUM, the investment policy statement, life stage, and prior meeting items. The workflow organizes account-level and household-level returns against benchmarks across multiple periods and identifies performance contributors and detractors. It also compares current and target allocations, builds talking points and an agenda, and proposes follow-up actions.
Collects the client name, household members, account types, total AUM, target allocation, risk tolerance, constraints, life stage, last meeting date, and open action items; formats QTD, YTD, 1-year, 3-year, and since-inception portfolio returns, benchmark returns, and alpha; identifies asset classes or positions driving results, including the top three contributors and detractors; compares target and current allocations and flags drift above the typical 3%–5% IPS threshold; creates an agenda covering market context, performance, allocation, planning updates, and action items; suggests rebalancing, tax-loss harvesting, cash deployment or withdrawal planning, Roth conversions, beneficiary or estate-planning updates, and insurance reviews; produces a one-page review summary, benchmarked performance table, current-versus-target allocation pie chart, recommended action items, and meeting agenda.
- A wealth advisor preparing a quarterly review needs a concise view of account-level and household-level performance.
- An advisor conducting an annual checkup needs to revisit the IPS, life stage, and allocation drift.
- An advisor handling an ad hoc client meeting needs market context, planning questions, and dated next steps.
- An advisor identifying allocation drift needs structured talking points for rebalancing and cash planning.
- An advisor documenting a completed review needs action items, deadlines, meeting notes, and any IPS changes captured.
How do you install this skill?
- Before handling account, tax, household, or IPS information, define authorization, access scope, storage, retention, and data flows.
- Rebalancing trades, tax-loss harvesting, Roth conversions, and insurance suggestions may create regulatory or suitability risk; label them as drafts requiring professional approval and never execute automatically.
- Missing-data behavior, benchmark selection, performance calculations, abnormal-input handling, and file-generation failures are unspecified; require manual validation before delivery.
- Chinese-language, jurisdictional, and mainland-China network compatibility are not stated; verify reachability and compliance separately when overseas data connectors are involved.
In Cowork, open Settings → Plugins → Add plugin, paste https://github.com/anthropics/financial-services, and select Meeting Prep Agent from the marketplace list. The README also describes Claude Code marketplace installation for the repository, but it does not provide a dedicated installation command for this specific agent.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/anthropics/financial-services.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/plugins/agent-plugins/meeting-prep-agent/skills/client-review" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- meeting prep for [client],
- prep for [client name],
After installation, use a trigger such as “client review,” “meeting prep for [client],” “quarterly review,” “prep for [client name],” or “client meeting.” Provide the client context, account data, IPS details, historical meeting notes, and open action items needed to prepare the review materials.
What are this skill's strengths and limitations?
- Provides an end-to-end meeting-preparation workflow spanning context, performance, allocation, planning, and follow-up.
- Requires both account-level and household-aggregate performance views.
- Includes benchmark comparison, performance attribution, and allocation-drift checks.
- Encourages direct discussion of poor performance, dated next steps, documentation, and compliance review.
- The repository is licensed under Apache License 2.0.
- The SKILL.md does not specify a data connector, account system, or data-retrieval tool.
- It requests Word or PDF output and charts without documenting how those artifacts are generated.
- The 3%–5% drift range is described as typical; the applicable threshold must come from the client’s IPS.
- The source material provides no test suite or verified platform coverage for this skill.
- The repository states that its content is not investment, legal, tax, or accounting advice and requires qualified professional review.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Client Review Prep this page ✓ Anthropic · Official | 44 · Not recommended | ★ 40k | 1mo ago | Apache-2.0 |
| Digital Brain Personal OS | 38 · Not recommended | ★ 18k | 14d ago | MIT |
| Diligence Meeting Prep ✓ Anthropic · Official | 48 · Use with care | ★ 40k | 1mo ago | Apache-2.0 |
| Notion Meeting Intelligence ✓ OpenAI · Official | 38 · Not recommended | ★ 28k | 4mo ago | — |
| Product Manager Skills | 59 · Recommended | ★ 186 | 6mo ago | NOASSERTION |
How did FollowSkills review this skill?
The skill explicitly requests client account, AUM, IPS, tax, and household information, all sensitive financial data. The README states that outputs are staged for qualified-professional review and do not execute trades, reducing external-effect risk. However, it lacks least-privilege rules, data-flow disclosure, client confirmation, privacy handling, rollback, and concrete compliance controls, so points are deducted.
The workflow and output structure are internally coherent across performance, allocation, agenda, and action items. However, it provides no handling for missing data, unavailable benchmarks, inconsistent periods, calculation errors, or inability to create Word/PDF/chart outputs, and there is no skill-specific test evidence, so points are deducted.
The audience, scenarios, and trigger phrases are reasonably clear for quarterly, annual, and ad-hoc client meetings. Boundaries, input formats, jurisdictional limits, Chinese-language support, and environment or connector requirements are unspecified, leaving semantic-trigger and environment-fit evidence incomplete, so points are deducted.
The staged workflow, tables, and output checklist are readable and useful. Repository context supplies an Apache-2.0 license, installation guidance, CI validation, and contribution path. The skill itself lacks versioning, changelog, examples, FAQs, troubleshooting, named ownership, and an update path; assumptions such as a typical 3-5% threshold are not configured, so points are deducted.
The skill covers the principal content needed for a client review and asks for a summary, performance table, allocation chart, and action items. It does not define data acquisition, calculation methodology, chart generation, or file-delivery mechanics, and tax and investment recommendations require substantial professional review; static evidence cannot establish directly usable delivery, so points are deducted.
The source provides an auditable workflow, while repository context shows plugin validation, secret scanning, and read-only CI permissions. There are no skill-specific fixtures, representative outputs, third-party execution records, or independently checked calculations, so only limited static verifiability is justified.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →