Equity research on SEC filings: point-in-time fundamentals, deterministic valuation, an SEC MCP server, and optional AI analyst memos.
IVI Analysis values public companies using the financial statements they file with the SEC. It downloads a company's XBRL data from EDGAR, cleans it up into annual and quarterly line items, and runs several standard valuation methods on it: a discounted cash flow model, earnings power value, the Graham number, free cash flow yield, EV/EBIT, net current asset value and a tangible book floor.
Everything runs locally and is stored in SQLite. You don't need any paid data. Financials come from the SEC for free and prices come from Yahoo Finance. If you add an Anthropic or OpenAI key, there is also an optional research layer that reads filings and writes an analyst memo, but the valuation numbers never depend on it.
The first run for a company downloads its SEC data (usually 3 to 8 MB) and caches it. Output looks like this:
The models are conservative on purpose. They use a 10% discount rate and don't give credit for growth that isn't already visible in the filings, so for a lot of well known companies the estimates come out well below the share price. Treat them as a floor based on what the company has actually reported, not a price target.
A few things the data layer handles that raw XBRL doesn't:
- You can ask for a company's numbers as of any past date and get only what had been filed by then. Later restatements don't leak into the historical view, which matters if you backtest anything. - When a figure has been restated, the most recently filed value is used. - Share counts on filing cover pages are sometimes off by a factor of 1,000. These are checked against other sha
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/Ryan-Halper--ivi-analysis.html)
Stop your AI from making things up — it proposes, deterministic tools decide, every claim checked against ground truth with evidence. Grounded facts and context survive resets. Reverse engineering is
yzg1226/mcd-party-officer🍟 麦当劳门店派对成团调度 Skill — 把派对从「策划」推进到「办得成」：成团概率雷达 / 包场vs拼团 / 择日黄历 / 儿童营养指数 / 招募卡与邀请函
hujjun/mcd-saver-strategist麦当劳省钱点餐助手（麦门精算师）｜麦当劳优惠券组合优化 Skill：拉取麦当劳菜单与全部可用优惠券，算出最省的点餐组合，巨无霸套餐 ¥47 → ¥21.9（省 53%）。McDonalds MCP（麦当劳 MCP）coupon optimizer · McDonalds China MCP Server · Agent Skill for WorkBuddy · 麦当劳程序员创意开发大赛参赛作品
sosoj92/jarvis-assistant-vocalAssistant vocal local en francais : Claude ou Ollama (offline), domotique Hue, OBS, agenda, navigateur, appels Twilio, serveur MCP. Python.
Stupidoodle/swissdevjobs-cliSearch & apply to ~4,700 salary-transparent tech jobs across 7 countries (🇨🇭🇩🇪🇬🇧🇺🇸🇨🇦🇳🇱🇫🇷) from your terminal or AI agent — zero-dependency Python CLI + MCP server + Claude Code plugin
krmisystems/fantasy-football-managerESPN draft and lineup tools for Codex and other MCP clients, with browser sign-in, approval modes, per-team limits, and server monitoring.
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.