natoshikamoto/KamotoAgent

KAMOTO — Open-source autonomous market agent for researching, analyzing, and executing strategies across equities, crypto, and onchain markets. MEATLOAF 0x49f139bc399bf865aeb51e787f1826b0edf52f04

What it solves

KAMOTO provides a framework for autonomous market intelligence, allowing AI agents to research assets, form investment theses based on data, and execute trades across both traditional and onchain financial markets without requiring constant manual instruction.

How it works

The system uses a reasoning core that integrates market research, portfolio awareness, and a risk engine. It follows an "observe-reason-decide-execute-learn" loop. The agent gathers data on price structure and volatility, evaluates it against the current portfolio, and then routes instructions through modular execution adapters that can connect to brokerage APIs, decentralized exchanges (DEXs), or simulation environments.

Who it’s for

It is designed for developers and researchers exploring the intersection of AI agents and autonomous finance, specifically those looking to build programmable financial strategies.

Highlights

  • Modular Execution: Supports various adapters for onchain protocols, brokerage APIs, and paper-trading.
  • Autonomous Research: Gathers structured information to build theses from multiple market signals.
  • Configurable Risk Controls: Allows users to set hard limits on position size, daily loss, and allowed assets.
  • Agent Memory: Preserves previous decisions and observations to provide context for future trades.
  • Experimental Market Primitives: Explores the ability for agents to create and interact with tokenized market primitives based on their own theses.

Related

  • Project
    lyon-industries/graphrag-workbench

    A local workbench that turns documents into an inspectable 3D knowledge graph by wrapping Microsoft GraphRAG with a visual interface for indexing and exploration.

  • Project
    jlumbroso/passage-of-time-mcp

    An MCP server that provides LLMs with temporal awareness and time calculation tools to accurately determine time differences and understand human temporal contexts.

  • Project
    offercontext/offerPilot

    An open-source, local-first AI job application tracker that helps job seekers manage resumes, track application progress, and prepare for interviews and salary negotiations.

  • Project
    liveloveapp/hashbrown

    An open-source framework for building AI agents in the browser, providing tools to embed intelligence into React and Angular components via a consistent LLM provider API.