@MayankTalwar0
基本信息
An MCP server for analyzing startup financial health and generating metrics reports locally.
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"startup-finance-metrics": {
"command": "uvx",
"args": [
"startup-finance-mcp"
]
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表，不代表服务没有工具。请查阅项目文档了解支持的能力。
点击上方按钮,在临时沙箱中运行此服务器并实时获取工具列表。
Startup Finance Metrics is an MCP server for analyzing startup financial health locally. It ingests bank statements, Stripe/QBO exports, or pasted values, uses AI to categorize transactions, computes key metrics (net burn, runway, gross margin, CAC, LTV, etc.), and generates unified Markdown and HTML reports. It is designed for startup founders and operators who need fast, on‑machine financial analysis without sending sensitive data to external services.
Install uv (curl/powershell commands provided), then configure your MCP client (e.g., Claude Desktop or Claude Code) with the command uvx startup-finance-mcp. The server exposes two tools: computeFinancialMetrics for computing metrics per month, and generateFinancialReport for producing a comparative report across one or more months. For manual prompting, a standalone skill prompt is available in skills/SKILL.md.
No. The server runs entirely locally, processes inputs in memory, and never sends data to any external API, cloud provider, or third‑party service (including SlickBooks). No data is stored, cached, or logged.
The AI categorizes transactions into revenue, COGS, S&M, payroll, or G&A based on descriptions. This step can make mistakes (e.g., misclassifying a contractor payment), so you should always review categorizations before sharing results with investors.
For best results, supply a minimum of three months of bank statements and active user stats. Sample files are available in the `test/` folder. The server will return `insufficient_data` with `missing_inputs` if required data is missing or ambiguous.
The server requires `uv` (a fast Python manager). Python does not need to be pre‑installed – `uv` automatically downloads everything needed in the background. It works with Claude Desktop, Claude Code, Glama, Cursor, and other MCP clients.
It calculates 12 metrics including Net Burn, Runway, Gross Margin, CAC, LTV, LTV:CAC, Revenue Growth, Logo Churn, Burn Multiple, NRR, Rule of 40, and CAC Payback. Each metric’s formula and required inputs are listed in the README’s metrics reference table.
Ask Claude or ChatGPT what Shopify theme optimization tasks are pending on your store, a remote MCP server for TaskerArmy Agent accounts.
Automated codebase documentation for GitHub: searchable docs, MCP context for AI agents, doc drift detection, and Friday Cleanup PRs.
Public read-only MCP server for discovering, inspecting, installing, and sharing approved Codex-compatible animated pet packs.
Query your team's drift, vulnerability, and migration data from any AI assistant. Vibgrate MCP connects Cursor, Claude, ChatGPT, Windsurf, or VS Code to Vibgrate Cloud: 51 tools for DriftScores, CVEs and EOL runtimes, up
Free enterprise-grade due diligence for the app you built, run by your own coding agent: 118 read-only checks plus an 18-dimension review, scored 1-5 with file-and-line evidence and a report page behind an access code. N
评论