selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

HunterCode Community Edition logo

Self-host HunterCode Community Edition

v2.2.1

Self-hosted AI investment research terminal for stocks

ReleaseOpen SourceApache-2.0
Updated2 weeks ago
LanguagePython
Setup difficultyModerate
Stars
576
CPU
2 CPU
RAM
4 GB
Storage
10 GB
Ports
3901 (OpenCode chat engine (internal)), 3999 (LLM shim (schema sanitizer)), 7233 (Temporal gRPC (fin profile, localhost only)), 8300 (Fin worker HTTP (fin profile, localhost only))
Requires
PostgreSQL, Redis
Install with Docker Compose

What is HunterCode Community Edition?

HunterCode Community Edition is a self-hosted AI investment research terminal designed for individual investors, private fund researchers, and small quant teams who want full control over their data. It covers A-shares, Hong Kong, and US stocks with real-time quotes, financial reports, news, and institutional fund flows. The system uses a multi-agent architecture: a chat interface powered by OpenCode (customized), an API backend (FastAPI), and an LLM shim for schema sanitization. Users provide their own LLM API keys (any OpenAI-compatible gateway, DeepSeek, Anthropic, or the built-in free quota via a hunt_tools_ platform key) and choose data sources (free AKShare/yfinance, the platform's data pipeline, or custom MCP servers). Core capabilities include UZI deep analysis (multi-dimensional reports), Kronos trend prediction (Tsinghua time-series model), a full-market screening filter, the Little Deer research desk, quantitative factor backtesting, and a paper-trading ledger (simulated trading only, no live broker connection). All data — chats, watchlists, positions, investment theses — stays in a local PostgreSQL database. Deployment is via Docker Compose with pre-built multi-arch images (amd64/arm64) from GHCR; a desktop launcher (HunterLauncher) automates the entire setup on macOS, Windows, and Linux. The project is licensed Apache 2.0.

Should you self-host HunterCode Community Edition?

HunterCode Community Edition scores 53/100 on selfhost.directory (how scoring works) and ranks #5 of 9 in Fintech. It has 576 GitHub stars.

It shipped 10 stable releases in the past year. The latest, v2.2.1, came out on 14 October 2026.

Setup is moderate: expect some configuration beyond a single command. It is Apache-2.0-licensed and written mostly in Python.

Its closest alternative on selfhost.directory is bitcoind (87/100, 90 527 stars). It scores higher overall, so compare the two before you pick. See HunterCode Community Edition vs bitcoind side by side.

Features

  • BYOK (Bring Your Own Key) for LLMs
  • Custom MCP server integration
  • Desktop launcher (HunterLauncher)macOS verified; Windows not yet tested on real hardware
  • Full-market screening & filtering
  • Graphical first-run wizard
  • Investment thesis tracking
  • Kronos trend predictionRequires platform key or self-hosted GPU
  • Little Deer research desk
  • Local-first data storage
  • Multi-agent investment research
  • Multi-user mode with registration policies
  • Paper trading ledger (simulated only)Opt-in via --profile fin; A-shares only
  • Real-time market data (A/HK/US)
  • SKILL system (Anthropic Agent Skills format)
  • UZI deep analysis

HunterCode Community Edition README

How to install HunterCode Community Edition

Install HunterCode Community Edition with Docker Compose

Environment variables
WEB_HOST_PORT, API_HOST_PORT, POSTGRES_HOST_PORT, REDIS_HOST_PORT, HUNTER_VERSION, HUNTER_REGISTRY, JWT_SECRET, HUNTER_SETUP_TOKEN, HUNTER_SINGLE_USER, REGISTRATION_MODE, LLM_BASE_URL, LLM_API_KEY, LLM_DEFAULT_MODEL, HUNTER_API_KEY, DATA_SOURCE_PROVIDER, FORECAST_PROVIDER, etc.

git clone https://github.com/agentpit-io/hunter-community cd hunter-community docker compose up -d open http://localhost:3100

Pre-built multi-arch images from GHCR; .env optional since v1.1.0

HunterCode Community Edition releases & version history

More in Fintech

Rate this project

Click to rate — no account needed.

0 comments

Posting as guest
Loading comments…