Build a full financial model without leaving Claude. The Layerz MCP plugin lets your agent spin up a structured model from a prompt, read values mid-conversation, edit formulas and scenarios, and export a clean .xlsx — your prompts, your tokens, your data.
MCP-compatible · BYOA · Export to standard .xlsx
You use Claude. You build financial analyses in chat. The next session, you start over.
You paste numbers, re-explain assumptions, rebuild context. Google Sheets can hold data. Airtable can organize records. PowerQuery can transform tables. None of them give your agent a structure it can navigate — typed variables, formula logic, timelines.
Layerz stores the structure, not just the numbers.
Read model structure — Variable names, types, formulas, dependencies — exposed natively to your agent.
Query computed values — Revenue, EBITDA, cash flow, any output at any period — already computed by the engine.
Update assumptions — Change a growth rate or headcount line, and every dependent output recomputes automatically.
Export Excel — Trigger a clean .xlsx from your agent session. Your colleagues open a standard file.
Your agent navigates the model. You review the outputs. The structure persists across sessions.
A budget or deal model your Claude agent fills from a prompt — structured outputs, not a raw table. Change one assumption, all outputs update. Queryable via API.
A Claude artifact, a Google Data Studio report, or an Excel file that pulls live values from your Layerz model — revenue, EBITDA, unit economics, scenario outputs. No manual refresh.
Unit economics, capacity planning, sensitivity analysis. Define the structure once, iterate across Claude sessions without losing prior states or re-explaining the logic.
Run base / upside / downside across a shared model structure. Export any scenario to clean .xlsx — your colleagues get a standard file, they never see the backend.
| Feature | Layerz | Google Sheets | Airtable | PowerQuery |
|---|---|---|---|---|
| Dynamic calculations (formulas, dependencies) | ✗ | partial | ||
| Structured for AI navigation | ✗ cells only | ✗ records only | ✗ ETL only | |
| Variables with type + timeline | ✗ | ✗ | ✗ | |
| Query outputs from external tools | ✓ via API | ✓ via API | ✗ | |
| Financial model logic (DAG, patterns) | ✗ manual | ✗ | ✗ | |
| Export to standard .xlsx | native | ✗ | native |
Layerz combines the dynamic calculations of a spreadsheet with the structure of a data model — built specifically for financial logic your agent can navigate without explanation.
Add to your Claude configuration (Desktop, Code, or any MCP-compatible client).
Sign in with OAuth — no API key copy-paste, no manual token rotation.
layerz_read, layerz_patch, layerz_get_model, layerz_build_from_blueprint, and more — invoked in conversation.
claude mcp add -t http layerz https://app.layerz.cc/mcpRun it, then `claude /mcp` to authenticate via OAuth.
| Client | Auth | Install |
|---|---|---|
| Claude Code | OAuth | claude mcp add -t http layerz https://app.layerz.cc/mcp |
| Claude Cowork | OAuth | claude.ai → Settings → Connectors → Add custom connector |
| Claude Desktop | OAuth | Settings → Connectors → Add custom connector |
| OpenAI Codex | OAuth | codex mcp add layerz --url https://app.layerz.cc/mcp |
| ChatGPT | OAuth | Settings → Apps & Connectors → Add new connector |
Layerz does not capture your Claude session. Connect your own Claude Max subscription or API key. Your prompts, your data, your environment.
Layerz is the structured backend, not the middleware.
Supported: Claude (MCP), any MCP-compatible agent, Claude Code, OpenAI Codex.
20+ finance professionals are using Layerz to build what their colleagues can’t explain — dashboards that update live, models their agents navigate without re-explanation, analyses that run in minutes instead of days. They’re not waiting for their company to approve a tool.
€0 /month
MCP plugin + REST API access — BYOA-compatible
€49 /month excl. tax
€89Unlimited projects · 30-day versioning · Layerz Share — no watermark
MCP-compatible · BYOA · Export to standard .xlsx