Okama Mcp
Let your coding agent run portfolio backtests, Monte Carlo simulations, and efficient-frontier analysis with chart PNGs while you build fintech or personal-finance features.
Overview
io.github.mbk-dev/okama-mcp is an MCP server for the Grow phase that exposes okama portfolio analytics—backtests, Monte Carlo, efficient frontier, and PNG charts—to coding agents.
What is this MCP server?
- Backtests and Monte Carlo via the okama library exposed as MCP tools
- Efficient frontier computation for allocation comparisons
- PNG chart generation for agent-readable visual outputs
- PyPI package okama-mcp v1.4.0 with uvx stdio transport
- MCP integration for Claude Code and Cursor—not a standalone trading app
- Server version 1.4.0 on PyPI package okama-mcp
- Stdio transport with runtimeHint uvx
- Capabilities: backtests, Monte Carlo, efficient frontier, PNG charts per server description
What problem does it solve?
Running portfolio backtests and risk simulations usually means leaving the agent to juggle Python snippets and chart code instead of callable tools.
Who is it for?
Indie builders prototyping fintech analytics, writing investment content, or validating allocation ideas with agent-driven quant steps.
Skip if: Teams needing licensed investment advice, live execution, or enterprise portfolio management without Python/uvx setup.
What do I get? / Deliverables
After you register the server, your agent can request standardized analytics and chart outputs you can paste into docs, decks, or product prototypes.
- Backtest and Monte Carlo results from agent-invoked tools
- Efficient frontier analysis outputs
- PNG chart files suitable for docs or UI mocks
Recommended MCP Servers
Journey fit
Portfolio performance and risk analytics belong in the grow phase when solo builders measure outcomes and refine product or investment narratives with data. Analytics subphase fits ongoing quantitative review—backtests and frontier plots are measurement workflows, not greenfield build tasks.
How it compares
MCP quant toolkit backed by okama, not a hosted robo-advisor or spreadsheet add-in.
Common Questions / FAQ
Who is io.github.mbk-dev/okama-mcp for?
Solo developers and small teams who use Claude Code or Cursor and want portfolio backtests, Monte Carlo, and efficient-frontier charts through MCP tools.
When should I use io.github.mbk-dev/okama-mcp?
Use it during grow-phase analytics or validate-phase pricing scenarios when you need repeatable portfolio metrics and PNG visuals inside an agent session.
How do I add io.github.mbk-dev/okama-mcp to my agent?
Add the PyPI package okama-mcp (v1.4.0) with uvx stdio transport to your MCP config per the server schema, then restart the client.