alphasift

Full-market stock discovery for AI agents: A-shares plus US/HK/JP/KR/TW/IN, auditable YAML strategies, optional LLM ranking, sector-diversified picks, and true

What is alphasift?

Full-market stock discovery for AI agents: A-shares plus US/HK/JP/KR/TW/IN, auditable YAML strategies, optional LLM ranking, sector-diversified picks, and true T+N evaluation vs. benchmark. AI选股

Overview of alphasift

What it is

AlphaSift is an open-source, full-market stock discovery tool designed for AI agents. It scans every stock in the China A-share market and supports six additional markets (US, Hong Kong, Japan, South Korea, Taiwan, India) in beta. The tool screens stocks using auditable YAML-based strategies, optionally ranks the shortlist with an LLM, ensures sector diversification in final picks, and saves every run for later T+N evaluation against a benchmark. It is licensed under Apache-2.0 and is available on GitHub.

Key features

  • Scans the full universe of A-shares (5,000+ names) and supports US, HK, JP, KR, TW, and IN markets via Yahoo Finance screener.
  • Strategies are defined in YAML files (filters, factor weights, scoring curves, risk thresholds, sector buckets) that can be diffed and reviewed.
  • Optional LLM ranking (Gemini, OpenAI, DeepSeek) re-ranks the shortlist after deterministic screening.
  • Sector diversity cap (portfolio_profile.mode: hard_cap) limits the number of picks per sector in the Top N.
  • Data sources are health-aware: dead sources are tried last, each source has a time budget, and a last-good snapshot is used when all live sources fail.
  • Every run can be saved and later evaluated with transaction costs, win rates, failure samples, and optional price paths (max drawdown/run-up).
  • Provides a read-only JSON API via alphasift serve for dashboards and orchestrators.

Use cases

  • AI agents performing automated, repeatable stock screening across multiple markets.
  • Quantitative researchers who want to define and backtest YAML-based strategies without writing code.
  • Investors seeking sector-diversified stock picks with transparent, auditable logic.
  • Developers integrating stock discovery into dashboards or orchestration pipelines via the JSON API.

Getting started

bash pip install -e . # A-shares, no API key needed alphasift screen dual_low --no-llm # Other markets: us, hk, jp, kr, tw, in alphasift screen global_value --market hk --no-llm # Add LLM ranking: set GEMINI_API_KEY / OPENAI_API_KEY / DEEPSEEK_API_KEY # (or LITELLM_MODEL / LLM_CHANNELS) in .env first cp .env.example .env alphasift screen dual_low # Save a run, then evaluate it a few days later alphasift screen dual_low --no-llm --save-run alphasift evaluate --with-price-path --explain Run alphasift quickstart for a guided no-key demo, and alphasift strategies to list the strategies.

License and pricing

The project is licensed under Apache-2.0 and is open source. No pricing information is stated beyond the license.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/zhulinsenalphasift.svg)](https://yunodb.com/tools/zhulinsenalphasift?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/zhulinsenalphasift?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/zhulinsenalphasift.svg" alt="Featured on YunoDB" height="20"></a>