Repolode
Submit
BruceLanLan

BruceLanLan/augur

Local-first investment research memory system where 18 investor personas analyze a stock independently, then combine into a weighted consensus with traceable evidence.

View on GitHubWebsite
Stars
681
Forks
115
Open issues
3
Last commit
9 days ago

Repolode describes open-source projects and does not give investment advice. Trading and investing can lose money: assess the risk yourself and follow the laws where you live before using this project.

What is it?

Augur is a local research tool for stock analysis. Eighteen rule-based investor personas, modeled on public investing philosophies such as value, growth, macro and China-market styles, each analyze a ticker independently. A weighted consensus engine then combines their signals, scores, confidence levels and Kelly position sizing, and shows where the personas agree and where they disagree.

Each piece of evidence carries its source and separate business, availability and retrieval timestamps. Missing data is marked explicitly, and personas that depend on it abstain with a reason rather than silently treating it as zero. Every run is stored as an immutable RunBundle that can be exported, compared and reviewed later. Data stays in a local directory by default, and the project includes no telemetry.

The project offers a CLI, a local web dashboard with pages for valuation DCF, earnings queues, thesis tracking and backtest evaluation, and an MCP server for connecting Claude or other agents. It pulls market data via yfinance and SEC EDGAR filings, and it is explicitly labeled as not investment advice.

Key features

  • 01Eighteen investor personas analyze the same ticker independently and produce a weighted consensus.
  • 02Evidence items record source and separate business, availability and retrieval timestamps.
  • 03Missing data is flagged explicitly, and dependent personas abstain with a stated reason.
  • 04Each run is saved as an immutable, exportable RunBundle for later review.
  • 05Includes a local dashboard, a CLI, and an MCP server for agent integration.
  • 06Pulls fundamentals from SEC EDGAR filings, such as 10-K risk-factor comparisons and XBRL deltas.

Use cases

  • Structuring a pre-earnings review of a watchlist of stocks.
  • Comparing how different value and growth investing frameworks view the same company.
  • Keeping an auditable history of past investment research runs and their assumptions.
  • Exposing a local research workflow to Claude or other agents through MCP.

Quick start

  1. 1

    Clone the repository

    git clone https://github.com/BruceLanLan/augur.git && cd augur
  2. 2

    Create a virtual environment and activate it

    python3 -m venv .venv && source .venv/bin/activate
  3. 3

    Install the core package with yfinance data support

    pip install -e ".[data]"
  4. 4

    Run the 18-persona analysis on a stock

    augur analyze AAPL

Things to know

  • Requires Python 3.10 or newer.
  • SEC EDGAR features need a real contact email set in AUGUR_EDGAR_CONTACT_EMAIL.
  • Investor personas are rule-based simulations of public investment ideas, not the actual investors' views, and the README warns it is not investment advice.

In collections

Open-source AI agent frameworks & appsMCP serversPrivacy & local-first tools

Topics

#investment-analysis#stock-analysis#multi-agent#ai-agents#local-first#sec-edgar#valuation#mcp#value-investing

FAQ

What is augur?
Local-first investment research memory system where 18 investor personas analyze a stock independently, then combine into a weighted consensus with traceable evidence. Augur is a local research tool for stock analysis. Eighteen rule-based investor personas, modeled on public investing philosophies such as value, growth, macro and China-market styles, each analyze a ticker independently. A weighted consensus engine then combines their signals, scores, confidence levels and Kelly position sizing, and shows where the personas agree and where they disagree.
Is augur free to use? What license does it have?
augur is open source under the MIT license. Check the license terms before using it in commercial products.
Is augur still maintained?
Actively maintained. The last commit to augur was on Oct 29, 2026.
What can I use augur for?
Structuring a pre-earnings review of a watchlist of stocks. Comparing how different value and growth investing frameworks view the same company. Keeping an auditable history of past investment research runs and their assumptions. Exposing a local research workflow to Claude or other agents through MCP.
How do I install augur?
Clone the repository: git clone https://github.com/BruceLanLan/augur.git && cd augur Create a virtual environment and activate it: python3 -m venv .venv && source .venv/bin/activate Install the core package with yfinance data support: pip install -e ".[data]" Run the 18-persona analysis on a stock: augur analyze AAPL
Where can I get augur?
The source code and installation instructions are on GitHub: https://github.com/BruceLanLan/augur

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects