explainx / blog
Anthropic's financial-services repo offers named agents (Pitch Agent, GL Reconciler, Market Researcher) and vertical plugins for investment banking, equity research, PE, and wealth management—9.5k GitHub stars, 11 MCP integrations.
May 7, 2026
At a San Francisco developer event, Anthropic rolled out Dreaming (research preview) to let Claude Managed Agents learn from session transcripts, plus public betas for multiagent orchestration (up to 20 parallel agents), outcomes loops (rubric-based self-checks), and webhooks for external notifications.
Jul 1, 2026
Anthropic''s official "Redeploying Fable 5" post — export controls lifted June 30, global restore July 1, stricter classifiers, 50% of weekly limits through July 7, credits after. HN breakdown: coding confusion, Opus fallback billing, invisible competitive-use limits, and Commerce reevaluation risk.
Jul 1, 2026
App strings showed Fable credits gated on verification before restore. Fable 5 is live July 1 — how credits and ID checks may still work.
Claude for Financial Services is Anthropic's open-source repository of reference agents, skills, and data connectors for FSI workflows—investment banking, equity research, private equity, and wealth management. With 9.5k GitHub stars, it offers 10 named agents (Pitch Agent, GL Reconciler, Market Researcher), vertical skill bundles, partner integrations with LSEG and S&P Global, and 11 MCP data connectors.
Everything is available two ways from one source: install as a Claude Cowork plugin, or deploy via Claude Managed Agents API behind your own workflow engine. Same system prompt, same skills—you choose where it runs.
Primary sources
| Topic | Details |
|---|---|
| What it is | Open-source reference agents, skills, and MCP connectors for FSI workflows |
| Named agents | 10 workflow agents: Pitch Agent, Market Researcher, GL Reconciler, Earnings Reviewer, Model Builder, etc. |
| Vertical plugins | 9 bundles: financial-analysis (core), investment-banking, equity-research, private-equity, wealth-management, fund-admin, operations, lseg, sp-global |
| MCP integrations | 11 providers: Daloopa, Morningstar, S&P Global, FactSet, Moody's, MT Newswires, Aiera, LSEG, PitchBook, Chronograph, Egnyte |
| Deployment | Cowork plugins (zip upload) or Managed Agents API (headless deployment) |
| File format | Markdown + YAML—no build step |
| GitHub stats | 9.5k stars, 1.3k forks, 7 contributors, Apache 2.0 license |
Nothing in this repository constitutes investment, legal, tax, or accounting advice. These agents draft analyst work product—models, memos, research notes, reconciliations—for review by a qualified professional. They do not:
Every output is staged for human sign-off. You are responsible for verifying outputs and for compliance with the laws and regulations that apply to your firm.
Each agent is named for the workflow it runs. They're starting points—install the ones that match your work, then tune prompts, skills, and connectors to how your firm does it.
Each agent plugin is self-contained—it bundles the skills it uses, so installing the agent is all you need.
| Agent | What it does |
|---|---|
| Pitch Agent | Comps, precedents, LBO → branded pitch deck, end to end |
| Meeting Prep Agent | Briefing pack before every client meeting |
| Agent | What it does |
|---|---|
| Market Researcher | Sector or theme → industry overview, competitive landscape, peer shortlist |
| Earnings Reviewer | Earnings call + filings → model update → note draft |
| Model Builder | DCF, LBO, 3-statement, comps—live in Excel |
| Agent | What it does |
|---|---|
| Valuation Reviewer | Ingests GP packages, runs valuation template, stages LP reporting |
| GL Reconciler | Finds breaks, traces root cause, routes for sign-off |
| Month-End Closer | Accruals, roll-forwards, variance commentary |
| Statement Auditor | Audits LP statements before distribution |
| Agent | What it does |
|---|---|
| KYC Screener | Parses onboarding docs, runs the rules engine, flags gaps |
For Managed Agent deployment—agent.yaml, leaf-worker subagents, steering-event examples, and per-agent security notes—see managed-agent-cookbooks/.
Start with financial-analysis—it carries the shared modeling skills and all data connectors. Add verticals for the workflows you need.
| Plugin | What it adds |
|---|---|
| financial-analysis (core) | Comps, DCF, LBO, 3-statement, deck QC, Excel audit. All 11 data connectors |
| investment-banking | CIMs, teasers, process letters, buyer lists, merger models, deal tracking |
| equity-research | Earnings notes, initiations, model updates, thesis and catalyst tracking |
| private-equity | Sourcing, screening, diligence checklists, IC memos, portfolio monitoring |
| wealth-management | Client reviews, financial plans, rebalancing, reporting, TLH |
| fund-admin | GL recon, break tracing, accruals, roll-forwards, variance commentary, NAV tie-out |
| operations | KYC document parsing and rules-grid evaluation |
| lseg (partner) | Bond RV, swap curves, FX carry, options vol, macro-rates monitoring on LSEG data |
| sp-global (partner) | Tear sheets, earnings previews, funding digests on S&P Capital IQ |
All connectors are centralized in the financial-analysis core plugin and shared across the rest.
| Provider | MCP URL | Access |
|---|---|---|
| Daloopa | https://mcp.daloopa.com/server/mcp | Subscription required |
| Morningstar | https://mcp.morningstar.com/mcp | Subscription required |
| S&P Global | https://kfinance.kensho.com/integrations/mcp | API key required |
| FactSet | https://mcp.factset.com/mcp | Subscription required |
| Moody's | https://api.moodys.com/genai-ready-data/m1/mcp | API key required |
| MT Newswires | https://vast-mcp.blueskyapi.com/mtnewswires | Subscription required |
| Aiera | https://mcp-pub.aiera.com | API key required |
| LSEG | https://api.analytics.lseg.com/lfa/mcp | Subscription required |
| PitchBook | https://premium.mcp.pitchbook.com/mcp | Subscription required |
| Chronograph | https://ai.chronograph.pe/mcp | API key required |
| Egnyte | https://mcp-server.egnyte.com/mcp | Subscription required |
Note: MCP access may require a subscription or API key from the provider. Check with your data vendor for authentication credentials.
In Cowork, open Settings → Plugins → Add plugin and either:
https://github.com/anthropics/claude-for-financial-services — then pick the agents and verticals you want from the marketplace list, orplugins/ (e.g., plugins/agent-plugins/pitch-agent/) and drop it in.# Add the marketplace
claude plugin marketplace add anthropics/claude-for-financial-services
# Core skills + connectors (install first)
claude plugin install financial-analysis@claude-for-financial-services
# Named agents — pick the ones you want
claude plugin install pitch-agent@claude-for-financial-services
claude plugin install gl-reconciler@claude-for-financial-services
claude plugin install market-researcher@claude-for-financial-services
# Vertical skill bundles
claude plugin install investment-banking@claude-for-financial-services
claude plugin install equity-research@claude-for-financial-services
Once installed:
/comps, /dcf, /earnings, /ic-memo)Each template under managed-agent-cookbooks/ references the same system prompt and skills as its plugin counterpart. The deploy script resolves file references, uploads skills, creates leaf-worker subagents, and POSTs the orchestrator to /v1/agents.
Deploy an agent:
export ANTHROPIC_API_KEY=sk-ant-...
scripts/deploy-managed-agent.sh gl-reconciler
Event routing:
See scripts/orchestrate.py for a reference event loop that routes handoff_request events between agents via your own orchestration layer.
Research preview note: Subagent delegation (callable_agents) is a preview capability. See per-agent READMEs for security and handoff guidance.
plugins/
agent-plugins/ # Named agents — one self-contained plugin each
vertical-plugins/ # Skill + command bundles by FSI vertical, plus MCP connectors
partner-built/ # Partner-authored plugins (LSEG, S&P Global)
managed-agent-cookbooks/ # Claude Managed Agent templates — one dir per agent
claude-for-msft-365-install/ # Admin tooling to provision Claude Microsoft 365 add-in
scripts/ # deploy-managed-agent.sh · check.py · validate.py · orchestrate.py · sync-agent-skills.py
| What it is | Where it lives |
|---|---|
| Agents | Self-contained plugins that own a workflow end to end—system prompt plus the skills it uses. Cowork and the Managed Agent wrapper both reference the same directory. |
| Skills | Domain expertise, conventions, and step-by-step methods Claude draws on automatically when relevant. Authored once in the verticals; each agent bundles a synced copy of the ones it needs. |
| Commands | Slash actions you trigger explicitly (/comps, /earnings, /ic-memo). |
| Connectors | MCP servers that wire Claude to your data—terminals, research platforms, document stores. |
| Managed-agent wrappers | agent.yaml + depth-1 subagents + steering examples for headless deployment. |
Everything is file-based—markdown and JSON, no build step.
Once vertical plugins are installed, these slash commands become available:
/comps — Comparable company analysis/dcf — Discounted cash flow valuation/lbo — Leveraged buyout model/3-statement — Integrated financial statement model/deck-qc — Quality check PowerPoint decks/excel-audit — Audit Excel models/cim — Confidential information memorandum/teaser — Teaser document/buyer-list — Generate buyer list/merger-model — Merger model/process-letter — Process letter/earnings — Earnings note/initiation — Initiation report/model-update — Update financial model/thesis — Investment thesis/catalyst — Catalyst tracking/ic-memo — Investment committee memo/diligence — Diligence checklist/sourcing — Sourcing pipeline/portfolio — Portfolio monitoring/financial-plan — Financial plan/rebalance — Portfolio rebalancing/client-review — Client review report/tlh — Tax-loss harvesting/gl-recon — General ledger reconciliation/accruals — Accruals analysis/variance — Variance commentary/nav-tieout — NAV tie-outLet's walk through using the Pitch Agent to generate a branded pitch deck from comps, precedents, and LBO analysis.
1. Install the agent:
claude plugin install pitch-agent@claude-for-financial-services
2. Trigger the agent in Cowork:
In Cowork dispatch, select Pitch Agent and provide:
3. The agent orchestrates:
4. Review and refine:
The agent stages the deck for human sign-off. You review, tweak assumptions, adjust formatting, and approve.
5. Export:
Final deck is exported as .pptx ready for client delivery.
Let's walk through using the GL Reconciler to find breaks, trace root cause, and route for sign-off.
1. Install the agent:
claude plugin install gl-reconciler@claude-for-financial-services
2. Trigger the agent:
Provide:
3. The agent orchestrates:
4. Review and sign-off:
The agent stages the reconciliation for human review. You verify root cause, approve adjustments, and close the period.
These are reference templates—they get better when you tune them to how your firm works.
Swap connectors: Point .mcp.json at your data providers and internal systems.
{
"mcpServers": {
"your-internal-data": {
"url": "https://api.yourfirm.com/mcp"
}
}
}
Add firm context: Drop your terminology, processes, and formatting standards into skill files.
## Your Firm's Comps Process
1. Pull universe from [internal screening tool]
2. Filter by [your firm's criteria]
3. Apply [your firm's multiples]
4. QC against [your firm's checklist]
Bring your templates: /ppt-template teaches Claude your branded PowerPoint layouts.
Adjust agent scope: Edit agents/<slug>.md to match how your team actually runs the workflow.
Add your own: Copy the structure for workflows we haven't covered.
If your firm runs Claude inside Excel, PowerPoint, Word, and Outlook via the Microsoft 365 add-in, claude-for-msft-365-install/ is the admin tooling to provision it against your own cloud—Vertex AI, Bedrock, or an internal LLM gateway—instead of Anthropic's API.
It's a Claude Code plugin (not a Cowork plugin) that walks an IT admin through:
Install:
claude plugin install claude-for-msft-365-install@claude-for-financial-services
/claude-for-msft-365-install:setup
This is separate from the agents and vertical plugins above—it's the on-ramp that gets the add-in deployed in a tenant, after which the agents and skills here are what runs inside it.
While Claude for Financial Services focuses on qualitative analysis (pitch decks, research notes, reconciliations), quantitative trading requires time series forecasting and backtesting.
Integration pattern:
Example workflow:
# Step 1: Generate sector overview
claude plugin run market-researcher --sector "semiconductor"
# Step 2: Get peer tickers → feed to Kronos for forecasting
# (Kronos batch prediction on NVDA, AMD, INTC, etc.)
# Step 3: Build DCF with Kronos forecasts
claude plugin run model-builder --type dcf --forecasts kronos_output.csv
# Step 4: After earnings → update models
claude plugin run earnings-reviewer --ticker NVDA --quarter Q1-2026
# Step 5: Generate investment memo
claude plugin run pitch-agent --target NVDA --deal-type "coverage initiation"
See our Kronos blog post for details on using Kronos for financial time series forecasting with Qlib integration.
Wealth-management vertical includes:
/financial-plan — Generate holistic financial plan/rebalance — Portfolio rebalancing recommendations/tlh — Tax-loss harvesting opportunities/client-review — Client review reportsIntegration with portfolio optimization:
/financial-plan to generate client goals and risk tolerance/rebalance to generate rebalancing recommendations based on optimizer output/tlh to identify tax-loss harvesting opportunities/client-review to generate quarterly client reportsEverything here is markdown and YAML. Fork, edit, PR. For new content:
New skill: Add it under plugins/vertical-plugins/<vertical>/skills/, then run:
python3 scripts/sync-agent-skills.py
This propagates to any agent that bundles it.
New agent: Create:
plugins/agent-plugins/<slug>/ (with agents/<slug>.md + skills/)managed-agent-cookbooks/<slug>/Pre-push check:
python3 scripts/check.py
This lints every manifest, verifies all cross-file references resolve, and fails if any bundled skill has drifted from its vertical source.
| Metric | Value |
|---|---|
| Stars | 9.5k |
| Forks | 1.3k |
| Watchers | 101 |
| Contributors | 7 |
| Language | Python (87.3%), Shell (7.2%), JavaScript (5.5%) |
| License | Apache 2.0 |
Key contributors: cxyback, manar-ant, mihilmy, tobinsouth, vinitra-ant, viveknair, alexa-perlov
Claude for Financial Services is Anthropic's open-source reference for FSI workflows—10 named agents (Pitch Agent, GL Reconciler, Market Researcher), 9 vertical plugins (investment-banking, equity-research, private-equity, wealth-management, fund-admin, operations), and 11 MCP integrations with Daloopa, FactSet, S&P Global, Moody's, and more.
Available two ways:
deploy-managed-agent.shUse it for:
The real power: These are starting points. Swap connectors, add firm context, bring your templates, and tune to how your firm works. Everything is markdown + YAML—no build step.
Get started:
# Claude Code
claude plugin marketplace add anthropics/claude-for-financial-services
claude plugin install financial-analysis@claude-for-financial-services
claude plugin install pitch-agent@claude-for-financial-services
# Or browse: https://github.com/anthropics/financial-services
Integrate with quantitative tools: Combine with Kronos for time series forecasting, or your own portfolio optimization models, for end-to-end investment workflows.
Nothing in this repository constitutes investment, legal, tax, or accounting advice. These agents draft analyst work product for review by qualified professionals. They do not make investment recommendations, execute transactions, bind risk, post to ledgers, or approve onboarding. Every output is staged for human sign-off.
You are responsible for verifying outputs and for compliance with applicable laws and regulations.
This article is an independent summary for developers on explainx.ai and is not sponsored by Anthropic. Features and specifications are based on the public GitHub repository as of May 7, 2026; verify on the official repo before production use.