</>MCP Agents Market
Agent

DojoAgents

by Alpha-Dojo3.1kPythonUpdated 2026-09-01

DojoAgents: Full-Market AI Copilot for Personal Investment

DojoAgents is an autonomous AI agent framework purpose-built for personal investment portfolio management across global markets (US, Hong Kong, and A-shares). It features a loop-driven cognitive engine that continuously analyzes market data, performs cross-market strategy deduction, and diagnoses portfolio risk exposure. The system combines a React-based dashboard with multi-agent collaboration to deliver institutional-grade market research, sector rotation tracking, and portfolio recommendations. Developers can deploy it locally, customize quantitative skills, and integrate their preferred LLM provider through a graphical configuration interface.

Key Features

Loop-driven cognitive portfolio agent with autonomous multi-turn reasoning and tool orchestration across four analysis dimensions (fundamental data cognition, advanced logic analysis, cross-market strategy deduction, and dynamic portfolio management)
Cross-market coverage spanning US equities, Hong Kong stocks, and A-shares with unified data ingestion and analysis
Interactive React dashboard featuring portfolio command center, market heatmaps, sector taxonomy drill-downs, and integrated K-line/PE band/financials views
Multimodal portfolio analysis supporting screenshot uploads for automatic holdings recognition and risk diagnosis
Autonomous background research agents that track sector rotations, monitor risk exposure, and conduct market surveillance while inactive
Extensible harness architecture with custom tool/skill directories, external SessionStore/BlobStore adapters, and plugin system for proprietary data integration
Secure execution sandbox for on-the-fly Python code execution, technical indicator calculations, and local web data extraction with strict financial hallucination guardrails
Multi-provider LLM support with in-app configuration for OpenAI, Anthropic, Google Gemini, Zhipu GLM, DeepSeek, and custom local endpoints (Ollama, llama.cpp, vLLM)

Use Cases

  • 01Daily cross-market briefing: automatically gather global market movers, sector strength rankings, and volume anomalies across A-shares, US equities, and HK stocks each morning
  • 02Breaking news impact analysis: upload a news headline (e.g., Meta infrastructure changes) and receive automated chain-of-reasoning linking affected stocks, sectors, and supply chains
  • 03Portfolio risk diagnosis: upload a holdings screenshot and receive concentration risk alerts, overlapping exposure warnings, and defensive asset recommendations
  • 04Simulated portfolio tracking: generate optimized watchlists from agent recommendations and monitor them continuously in the dashboard with live net-worth curves
  • 05Sector rotation monitoring: deploy background agents to track momentum shifts across L1/L2/L3 sector taxonomy and identify emerging allocation opportunities
  • 06Quantitative workflow automation: distill successful multi-step market analysis sequences into reusable SKILLS for programmatic future execution

Related Agents

View more

DojoAgents — FAQ

What is DojoAgents?+

DojoAgents is an autonomous AI agent framework for personal investment that uses a loop-driven cognitive engine to analyze global markets, diagnose portfolio risk, and perform cross-market strategy deduction. Unlike simple financial AI tools that provide quick summaries, it features multi-agent collaboration and portfolio-aware reasoning across US, Hong Kong, and A-share markets.

How do I install DojoAgents?+

Install via PyPI using 'uv pip install dojoagents' (or 'pip install dojoagents') after creating a Python 3.11+ virtual environment. Developers building from source should clone the repository, run 'uv pip install -e .[dev]', then build the React dashboard with 'npm install && npm run build' in the dojoagents/dashboard/web directory. Launch the server with 'dojoagents dashboard --host 127.0.0.1 --port 8765'.

Which AI clients or platforms does DojoAgents work with?+

DojoAgents is a standalone agent framework with its own FastAPI backend and React dashboard, not a plugin for existing AI clients. It connects to LLM providers (OpenAI, Anthropic, Google Gemini, Zhipu GLM, DeepSeek, or custom local endpoints) through its in-app Settings interface, with future planned support for delivering insights via Slack, Telegram, Discord, Feishu, WeChat, and email.

Do I need API keys or prerequisites to use DojoAgents?+

Yes, you need an API key from at least one LLM provider (OpenAI, Gemini, Anthropic, etc.) to power the agent's reasoning engine. You also need Python 3.11+, Node.js 18+, and npm 9+ if building the dashboard from source. All API keys are configured securely through the graphical Settings interface and stored locally in ~/.dojo/agents.yaml.

Is DojoAgents free to use?+

Yes, DojoAgents is open-source under the Apache License 2.0 and free to use. However, you will incur costs from your chosen LLM provider based on API usage (token consumption for market analysis and reasoning). The framework itself has no licensing fees.

Can I customize DojoAgents with my own quantitative tools or data sources?+

Yes, the harness architecture supports custom tool/skill directories that supplement the built-in financial harness. Developers can write custom plugins, integrate proprietary data sources, implement external SessionStore/BlobStore adapters for production deployments, and distill successful analysis workflows into reusable programmatic SKILLS. Full documentation is provided for plugin development and dashboard customization.

How do I install DojoAgents?+

Open the source repository on GitHub and follow its README. DojoAgents is a agent — MCP Agents Market links you directly to the official repo.

Is DojoAgents free?+

DojoAgents is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.

Related searches