Investo
Investo is an MCP server that provides an AI investment-analysis agent with structured financial data and analytical tools for Indian (NSE/BSE) and global companies.
Search & profile: Resolve company names to tickers and get sector, business summary, market cap, and executives.
Fundamentals: Retrieve income statements, balance sheets, cash flow, and key valuation/profitability/leverage/liquidity/growth ratios.
Peer & industry analysis: Compare competitors, view sector sub-domains, demand drivers, CAGR, and industry risks.
News & management: Get categorized news headlines, plus executive, promoter/insider holding, and capital-allocation insights.
Valuation: Run DCF with configurable discount/growth/years, see intrinsic value/share, margin of safety, and expected return.
Qualitative assessments: Evaluate economic moat, risk signals, a 0–100 investment score, and a Warren Buffett-style checklist.
Advanced analytics: View shareholding patterns, growth outlook, fundamental trends, red flags, investment thesis, technical snapshot, DCF sensitivity, and peer-relative metrics.
Comparison & reporting: Compare 2–6 companies head-to-head, explore curated peer groups, and generate full HTML/PDF research notes or a bundled all-in-one analysis.
Fetches categorized recent news headlines about companies from Google News.
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., "@Investo Analyze Infosys"
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.
Investo 📈
An AI investment-analysis agent you run from Claude or Cursor.
⚠️ Research and education only — not investment advice.
Give Investo a company name — Indian (NSE/BSE) or global — and it gathers public financial data and produces a full analysis: what the company does, its financials & ratios, a competitor comparison, DCF intrinsic value, economic moat, risks, management, recent news, SWOT seeds, and a 0–100 investment rating.
Investo is a Model Context Protocol (MCP) server. It exposes tools to an AI client (Claude Code, Claude Desktop, Cursor); the client calls those tools and writes the analysis narrative grounded in the structured data Investo returns.
Primary focus: Indian companies listed on NSE (
.NS) and BSE (.BO). US/global companies are supported too.
What it produces
For any company, Investo supplies the evidence for:
Domain / sector — what the business does and its sub-domains.
Financials & ratios — income statement, balance sheet, cash flow + valuation, profitability, leverage, liquidity, growth, cash-flow ratios.
Competitor analysis — auto-compares against sector peers (e.g. Infosys → TCS, Wipro, HCL, Tech Mahindra, LTIMindtree).
Industry intelligence — sub-domains, demand drivers, CAGR, risks.
News analysis — recent headlines categorized (earnings, M&A, management, legal, product/AI).
Management analysis — executives, promoter/insider holding, capital allocation.
DCF valuation — intrinsic value/share, margin of safety, expected return.
Economic moat — brand / network / cost / scale / switching-cost signals.
Risk analysis — debt, currency, concentration, regulation, tech obsolescence.
Rating out of 100 — a balanced 11-bucket score with per-bucket rationale.
Warren Buffett checklist — a weighted 0–100 quality-fit score; each criterion (ROE, ROIC, debt, owner earnings, margin of safety, management, moat) shows value vs threshold, a pass/warn/fail with the reason, a confidence, and its multi-year trend.
Relative to industry — key metrics vs the peer-set median with favourable-side percentiles.
Shareholding pattern — promoter/FII/DII/public split + promoter pledge, with quarter-over-quarter smart observations and an ownership signal (NSE/BSE filings; Yahoo fallback).
5-year growth engine — the primary engine plus ranked drivers (estimated contribution %, per-driver risks), a catalyst timeline, and a blended growth band.
Fundamentals trend, red-flags, and an investment thesis — multi-year health at a glance, automated deterioration warnings, and a synthesized pros/cons verdict.
Every section carries a confidence score, provenance and reasoning (the evidence layer), so an
AI agent — or you — can judge how far to trust each conclusion. A machine-readable ai_signals
digest and a self-contained, print-ready research note (--html) — numbered sections, inline
SVG exhibits, footnotes and per-exhibit source lines, styled as an institutional equity-research
document rather than a dashboard — are available too.
Rating buckets (out of 100)
Growth | Profitability | Cash Flow | Balance Sheet | Valuation | Moat | Management | Industry | Innovation | Risk | ESG* |
15 | 17 | 10 | 11 | 10 | 12 | 10 | 5 | 5 | 5 | 5* |
*ESG is optional; when unavailable the remaining buckets renormalize to 100.
Valuation is quality-aware. A premium multiple isn't punished when the company's economics justify it — the acceptable P/E · P/B · EV/EBITDA ceilings widen with ROE, margins and growth, so a proven compounder isn't floored just for not being cheap (while a low-quality expensive name still is). Net cash strengthens the score — it lifts the Balance Sheet bucket and lowers the effective equity multiple — and net debt weakens both.
Related MCP server: financial-research-agent
Install
Requires Python 3.10+.
git clone https://github.com/YashvantHange/Investo
cd Investo
python -m venv .venv
# Windows: .venv\Scripts\activate | macOS/Linux: source .venv/bin/activate
pip install -e .No API keys are required — Investo works out of the box using free Yahoo Finance data and
Google/Yahoo news. Optional keys (Alpha Vantage / FMP / Finnhub) enable richer/fallback data;
copy .env.example to .env and fill in any you have.
Try it from the command line
investo analyze "Infosys" # terminal report + auto HTML note
investo analyze "Reliance Industries"
investo analyze "Tata Motors"
investo analyze AAPL
investo analyze "Reliance Industries" --html reliance.html # self-contained research note
investo analyze "Infosys" --pdf infosys.pdf # PDF via headless Chrome/Edge
investo analyze "Infosys" --json --html infy.html # flags compose; nothing is discarded
investo analyze "Infosys" --no-html # skip the automatic HTML note
investo analyze "Infosys" --open # open the note in your browser
investo search "tata motors"Every investo analyze writes a self-contained HTML research note automatically (named
investo-<SYMBOL>-<date>.html in the working directory) alongside its terminal output — the path is
announced on stderr, so --json stays pipeable. Use --no-html to skip it, or --html FILE to
choose the location. Add --open to open the written note in your default browser. The MCP
analyze_company tool does the same, returning the file in html_report_path; pass
emit_html=false to suppress it. Because a chat client will refuse to follow a file:// link, the
tool also returns html_report_open_url — an http://127.0.0.1 link served by a loopback-only
static server, so the report opens on click.
--pdf needs a Chromium-family browser: it uses a system Chrome, Edge, Chromium or Brave if one
is installed (no setup), falls back to a managed Chromium via pip install 'investo[pdf]' && playwright install chromium, and otherwise prints exactly how to fix it while still leaving the
.html on disk. Point INVESTO_CHROME at a specific executable to override discovery. Bare --html
/ --pdf (no filename) write investo-<SYMBOL>-<date>.<ext> in the working directory.
Use it from Claude Code / Cursor
Do the one-time setup (creates the venv the launcher looks for):
python -m venv .venv
.venv\Scripts\pip install -e . # macOS/Linux: .venv/bin/pip install -e .Claude Code — this repo ships a project-scoped .mcp.json that runs
python scripts/mcp_launcher.py. No paths to edit — the launcher finds the project's
.venv itself and works on Windows/macOS/Linux. Opening the folder in Claude Code offers to
load the investo server (approve on first use); the included CLAUDE.md makes the agent
introduce itself as Investo.
Cursor — Investo is in the Cursor Directory. One-click
install (requires uv — the Python equivalent of npx):
Or add manually to .cursor/mcp.json (project) or ~/.cursor/mcp.json (global) — both work:
{ "command": "uvx", "args": ["--from", "git+https://github.com/YashvantHange/Investo", "investo-mcp"] }uvx builds & runs Investo straight from GitHub — no clone, no venv, works from any folder.
Claude Desktop — use the same uvx config (see examples/claude_desktop_config.json), or
install the one-click .mcpb bundle (scripts/build_mcpb.sh).
From source (no uv) — clone, python -m venv .venv && pip install -e ., then point the
MCP config at the launcher: { "command": "python", "args": ["<ABSOLUTE>/scripts/mcp_launcher.py"] }
(the launcher finds the venv itself). This is what a project-scoped .mcp.json uses.
See PUBLISHING.md for PyPI / .mcpb / MCP-registry release steps.
Then ask: "Analyse Infosys", "Compare HDFC Bank with its peers", "What's the DCF value of Reliance?"
How the launcher works:
scripts/mcp_launcher.pyis a tiny standard-library script. When a client runs it with anypython, it re-launches the server inside the project's.venv(or uses the current interpreter ifinvestois already installed there). That's why the committed config needs no machine-specific paths.
MCP tools
Tool | Purpose |
| Resolve a name to an NSE/BSE/global ticker |
| Sector, business summary, market cap, executives |
| Income statement / balance sheet / cash flow |
| Valuation, profitability, leverage, growth, cash-flow ratios |
| Competitor comparison table |
| Sub-domains, demand drivers, CAGR, risks |
| Categorized recent headlines |
| Executives, holdings, capital allocation |
| Intrinsic value, margin of safety, expected return |
| Economic-moat signals + heuristic score |
| Risk signals + heuristic score |
| 0–100 composite rating |
| Warren-Buffett quality checklist: weighted 0–100 fit, per-criterion pass/warn/fail + reason, confidence & multi-year trend |
| Key metrics vs the peer-set median (industry proxy) with favourable-side percentiles |
| Promoter/FII/DII/public split + pledge, QoQ smart observations & ownership signal (NSE/BSE filings, Yahoo fallback) |
| 5-year growth engine: ranked drivers (contribution %, risks), catalyst timeline, blended growth band |
| Multi-year revenue/profit/margin/EPS/ROE with per-year direction & health grade |
| Automated deterioration warnings + overall risk level |
| Synthesized pros/cons, quality grade, valuation stance & one-line verdict |
| Compact machine-readable digest (thesis, quality, confidence, ownership/growth signals, risk, valuation) |
| Price/momentum context: 50/200-DMA + golden/death cross, RSI, volatility, drawdown, beta, 52-week position (context, not a signal) |
| Intrinsic value across a discount-rate × terminal-growth grid + the growth implied by today's price |
| Head-to-head across 2–6 named tickers (not a curated group) |
| List the curated peer groups and their members |
| Render a full analysis to an HTML/PDF file (writes a file; path sandboxed) |
| Everything above bundled into one report (with a confidence/provenance evidence layer); also auto-writes an HTML note and returns its |
| SEC EDGAR cross-check (US/ADR only) |
Configuration
All optional — set as environment variables (or in .env; see .env.example):
Variable | Purpose | Default |
| Licensed data (primary when set) | — |
| Log verbosity to stderr (DEBUG/INFO/WARNING/ERROR) |
|
| Min seconds between Yahoo calls |
|
| Alpha Vantage daily cap before Yahoo fallback |
|
| Contact for the SEC EDGAR User-Agent | repo URL |
| Fetch NSE/BSE shareholding filings (else Yahoo fallback) |
|
|
|
|
| DCF discount / terminal / years overrides | see |
Data sources & legal
Investo prefers licensed data when you configure a key, and falls back to free Yahoo data otherwise:
With an API key (
ALPHAVANTAGE_API_KEY/FMP_API_KEY/FINNHUB_API_KEY): licensed fundamentals are used as the primary source and take precedence for the fields they cover (recommended for production / commercial use).Without a key (default, zero-config): Yahoo Finance is used via
yfinance, which relies on Yahoo's public but unofficial endpoints. This is best-effort, may be rate-limited, and is subject to Yahoo's terms of service. For NSE/BSE fundamentals Yahoo remains the practical source of record even when a key is set, because the licensed APIs' India coverage is limited.
The provider in effect is reported by the provider_status in tool output. See
SECURITY.md for the full list of endpoints Investo contacts.
Privacy — what leaves your machine
Only the company name or ticker you ask about is sent to the data endpoints above. Investo has no telemetry, stores no personal data, and reads API keys only from environment variables (never logged). It is read-only and does not modify your system.
Known limitations
Promoter/insider shareholding for NSE/BSE has no clean free API — best-effort, often unavailable for Indian names.
Industry CAGR / market share are curated/estimated (
data/*.yaml), not live. Each peer group carries anupdated_atso you can judge staleness rather than assume freshness.Peer lists start curated for major Indian sectors and are extensible via
data/peers.yaml. A ticker in no group falls back to a keyword match on its Yahoo industry; that guess is reported asbasis: sector-fallbackand scored below a curated group. After editing peers.yaml, runpython scripts/validate_peers.py— a dead ticker silently drops a company out of its own peer table, and no offline test can catch it.Confidence is about evidence quality, not about being right — see docs/confidence.md for how it's computed and where it stops being trustworthy.
Sharp reporting discontinuities (e.g. a demerger) can distort growth; Investo flags a warning when it detects one, but read the note in context.
⚠️ Investo is for research and education only — not investment advice. Do your own due diligence.
License
MIT
Available Tools
14 toolsanalyze_companyB
Full investment analysis for a company name or ticker.
Bundles profile, ratios, peer comparison, DCF, moat, risk, management, news, the 0-100 score, plus SWOT seeds and growth-driver hints for the host LLM to turn into a narrative.
Inputs: query (required), market.
compare_peersB
Competitor comparison table vs sector peers (revenue, margins, valuation, growth).
Inputs: ticker (required), market.
dcf_valuationB
Two-stage DCF: intrinsic value/share, margin of safety and expected return.
Optional overrides: discount_rate (e.g. 0.12), terminal_growth (e.g. 0.04), years, growth_rate. Handles cases where statements and the stock trade in different currencies.
Inputs: ticker (required), years, market, growth_rate, +2 more.
get_company_profileC
Company profile: sector, industry, business summary, market cap, executives.
Inputs: ticker (required), market.
get_financialsC
Income statement, balance sheet and cash flow (period = 'annual' or 'quarterly').
Inputs: ticker (required), market, period.
get_industry_intelligenceC
Sector sub-domains, demand drivers, industry CAGR and risks.
Inputs: ticker (required), market.
get_key_ratiosC
Valuation, profitability, leverage, liquidity, growth and cash-flow ratios.
Inputs: ticker (required), market.
get_managementC
Executives, promoter/insider/institutional holding and capital-allocation signals.
Inputs: ticker (required), market.
get_newsB
Recent company news, categorized (earnings, M&A, management, legal, product/AI).
Inputs: ticker (required), limit, market.
get_sec_factsB
SEC EDGAR company facts (US-listed companies / Indian ADRs only). Optional cross-check.
Inputs: ticker_or_cik (required).
moat_assessmentC
Economic-moat signals (brand/cost/scale/IP) with a 0-10 heuristic score.
Inputs: ticker (required), market.
risk_assessmentC
Risk signals (leverage, currency, concentration, regulation) with a 0-5 safety score.
Inputs: ticker (required), market.
score_companyC
The 0-100 investment rating with its eleven weighted buckets and rationale.
Inputs: ticker (required), market.
search_companyA
Resolve a company name to an exchange ticker (NSE/BSE preferred for India).
Returns the best match plus ranked alternatives.
Inputs: query (required), market.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 14 tools
Each tool targets a distinct aspect of investment analysis (profile, financials, ratios, valuation, moat, risk, etc.), with no overlapping purposes. The comprehensive description helps an agent clearly distinguish between them.
All tools follow a consistent verb_noun pattern (e.g., get_company_profile, analyze_company, dcf_valuation) using lowercase and underscores. This makes the tool set predictable and easy to navigate.
With 14 tools, the server covers the full breadth of investment analysis without being excessive. Each tool adds specific value, and the count matches the domain's complexity well.
The tool set covers all major areas of fundamental analysis: company lookup, profile, financials, ratios, peer comparison, DCF valuation, moat, risk, management, news, SEC data, and a scoring system. No obvious gaps for the intended purpose.
Maintenance
Related MCP Connectors
Hosted MCP server for Indian NSE/BSE stock research: 59 tools for quotes, financials and ratios, peers, shareholding and pledge, filings with extracted facts, screens, F&O and FII/DII context. Source and freshness on every answer; no buy/sell calls.
The Octagon MCP server provides specialized AI-powered financial research and analysis by integrating with the Octagon Market Intelligence API. It enables users to analyze public market data (SEC filings, earnings transcripts, financial metrics, and stock data for 8000+ companies), private market data (3M+ companies, 500k+ funding rounds, 2M+ M&A/IPO transactions), and conduct deep research including web scraping capabilities. The server also features autonomous research agents that search hundreds of sources and return fully cited reports in approximately one minute.
Official Intrinio MCP server. US stock prices, options and fundamentals for AI agents.
Research-only MCP server: your AI as a quant research desk. 90 tools, no trades, no brokers.
Related MCP Servers
- AlicenseAqualityDmaintenanceA comprehensive MCP server for stock analysis and trading insights, including stock screening, fundamental analysis, insider trading, options analysis, social media research, and news analysis.1075MIT
- FlicenseNot gradedqualityDmaintenanceMCP server that exposes stock research tools (fundamentals, news, technicals, analyst ratings) to AI clients, enabling autonomous generation of structured investment briefs.-
- AlicenseNot gradedqualityDmaintenanceAn AI-powered MCP server for value investors. Analyze stocks using principles from Benjamin Graham and Warren Buffett.9 npm25MIT
- AlicenseBqualityCmaintenanceComprehensive MCP server for Indian stock market with 60 tools covering NSE/BSE, derivatives, mutual funds, technicals, and more, requiring no API keys. Enables users to access live market data, analyze indicators, and manage investments through natural language.4601MIT