Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Fund Dashboard MCP Server Calculate the XIRR for my portfolio this year"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Personal investment portfolio analysis platform: fund / stock position management, net value tracking, XIRR annualized return, maximum drawdown, look-through analysis, DCA dollar-cost averaging backtesting, Monte Carlo simulation, and an MCP server for AI agents.
Pure Go backend + React frontend, single-container deployment, SQLite persistence (optional PostgreSQL).
Position management: unified management of funds + stocks (A-shares / HK stocks / US stocks), transaction CRUD, automatic snapshot recalculation
Quantitative analysis: XIRR annualized return, maximum drawdown, portfolio NAV trends, profit & loss distribution, Monte Carlo simulation, correlation heatmap, fund comparison radar, DCA backtesting
Look-through analysis: drills through fund holdings to the underlying stocks, aggregated by industry
MCP server: AI agent tools (portfolio / transactions / admin / market / analysis / report), Streamable HTTP transport, Bearer key authentication
Dual themes: light / dark theme, red for gains, green for losses
Related MCP server: FinanceKit MCP
Layer | Technology |
Backend | Go ( |
Frontend | React + Vite + echarts ( |
Contracts | zod schemas shared between frontend and backend ( |
Storage | SQLite (default), optional PostgreSQL dual-driver |
Auth | Bearer key (operator / analyst dual scope) + write-path key injected by edge proxy |
# 测试
go test ./... -count=1
npm ci && npm test
# 构建
go build -o bin/fund-dashboard ./cmd/fund-dashboard
npm run build --workspace packages/web
# 容器
docker build -f deploy/Dockerfile -t fund-dashboard:local .Document | Content |
Architecture | |
UI design system | |
Testing system | |
Changelog | |
Contribution and release process | |
Build and run | |
Agent shared constraints |
This project follows Semantic Versioning. Releases are triggered by Git tags:
./scripts/release.sh 2.0.0 # 归并 CHANGELOG + 打 tag v2.0.0 + 推送Once a tag is pushed, GitHub Actions automatically builds multi-architecture images (GHCR) and creates a Release, with release notes taken from CHANGELOG.md. To pull an image:
docker pull ghcr.io/deliciousbuding/fund-dashboard:latestIssues and PRs are welcome. Development workflow: create a feature/* branch from main, and open a PR back to main. See CONTRIBUTING.md.
License: Apache-2.0
This server cannot be deployed
Personal-finance workspace for AI agents: accounts, spending, budgets, goals, and investments.
Investment research superagent: podcasts, SEC filings, and no-code research pipelines.
Unified financial infrastructure connecting AI agents directly to trade live/demo brokerage accounts, Web3 non-custodial wallets, real-time market data across equities, ETFs, crypto, forex, options, DeFi swaps, and prediction markets, institutional research feeds, and algorithmic strategy backtesters.
Connect AI agents to bank accounts, transactions, balances, and investments.
