One command. 24 AI agents. A 股深度研究.

Turn a 6-digit stock code into a 16-agent deep-dive report — fundamentals · technicals · whale signals · quant scores · bull/bear/risk debate — in ~10 minutes.

English  ·  中文

What is it  ·  Features  ·  Quick Start  ·  Agents  ·  Memory  ·  Datasets  ·  LLM

🐣 New to Python / CLI? 小白上手指南 (中文, 30 min) →

🆕 v1.0.7 highlights (2026-05-26)

• fa data update 全开 5 种新数据 — --include-f10 (TDX 公司大事/龙虎榜/研究报告, 零 token, pytdx 直连) · --include-concepts (同花顺概念股 + 成分股, 零 token, adata) · --include-northbound (沪+深股通历史资金流向, 零 token, akshare) · --include-financial (Tushare 财务三表, opt-in) · --include-stock-basic (Tushare 公司基本信息, opt-in)
• UI 数据按钮 Shift+点击全开 — 普通点击 = 日线 + 5min + daily_basic + 北向 (~5 min 安全); Shift+点击 = 全开 (+ F10 csi500 ~30 min + 概念股 + Tushare 两项)
• buddy /data/refresh 7 个 query 跟 CLI 一一对齐 — UI 一键映射 CLI flag, Tushare token 走 server env FA_TUSHARE_TOKEN 不在 URL 暴露
• last_update.py 扩 8 个数据类型 — /data/status 现在追踪 day/5min/daily_basic/financials/f10/concepts/stock_basic/northbound 8 类 staleness
• 3–10× faster data downloads (v1.0.6 起) + ModelScope (魔搭) CN-CDN (FA_DATA_SOURCE=modelscope)

Full CHANGELOG.

A-share research workstation that thinks like a buy-side analyst.

Hand it a stock code; 14 specialized AI sub-agents run in 4 trust tiers:

Out comes a markdown research report — rated, attributed, falsifiable. The report-writer is the only agent allowed to write report files. Untrusted news/F10 sources are JSON-schema-locked at Tier-1 (no prompt injection). Memory is markdown — edit a .md, next report uses it. FTS5 retrieval cuts prompt cost ~60%.

fa report SH600519

fa brief

fa overseas-radar

fa mainline

fa dream --since 30

financial-analyst    # /model deepseek-reasoner

20 fa tools accessible from any AI IDE that speaks the Model Context Protocol.

• stdio — financial-analyst-mcp console script auto-installed by pip. Works with Claude Desktop, Claude Code.
• HTTP streamable — fa start auto-mounts the same tools at http://127.0.0.1:9999/mcp. Works with Cursor, Codex CLI, JetBrains AI plugins.
• Same 20 tools, two transports — read (quick_quote, memory_search, read_past_report, chain_lookup, ...) <1s; long (report, data_update) covered by the read_past_report workaround; dream-loop mutation tools (accept_proposal, revert_proposal) write to ~/.financial-analyst/audit.jsonl and git add the change so every memory edit is observable and reversible.

Full client config for all 4 IDEs → docs/mcp.md

pip install financial-analyst==1.0.7
fa start                   # interactive wizard (LLM key + workspace + HF dataset)
                           # then auto-starts backend + UI + browser
# or non-interactive:
fa init --yes --preset demo --workspace D:/fa-workspace   # CI / scripted
fa report SH600519                                         # first deep-dive (~10 min)

git clone https://github.com/jesson-hh/financial-analyst.git
cd financial-analyst
pip install -e ".[dev]"
pytest tests/              # 712 tests, ~8 min

Full DAG: docs/architecture/14_agents.md

memories/
├── README.md                        # ← directory index, must-read
├── risk-officer/
│   ├── hard_rules.md                # ← edit this → next report uses it
│   └── pitfalls.md                  # FTS5-retrieved (large file)
├── technical-analyst/
│   └── factor_insights.md
└── _shared/
    └── playbook_V1_V10.md           # cross-agent shared

Batteries included. A fresh pip install seeds ~/.financial-analyst/memories with 24 agents' worth of curated rules/playbooks — the same knowledge the maintainer runs: pitfalls, 440-factor insights, the V1–V10 analyst framework, sentiment signals, and the market-cap-tiered rating system (~4,300 lines). A working analyst brain on day one, not an empty shell. Seeding only fills an empty dir, so your edits and accumulated dream-loop memory are never overwritten on upgrade.

Edit a markdown → next agent run picks it up. No restart, no rebuild.

See memories/README.md for the 24 dir index and design principles.

Three preset bundles on HuggingFace, fa init auto-pulls:

from huggingface_hub import snapshot_download
snapshot_download(
    repo_id="yifishbossman/financial-analyst-data-lite",
    repo_type="dataset",
    local_dir="~/.financial-analyst/data",
)

Two binary formats: Qlib .bin (time-series, [4-byte float32 start_idx] + [float32 array]) for OHLCV+factors; Parquet (columnar) for financials/events/F10/industry. Compatible with Microsoft Qlib and D.features() API directly.

HuggingFace is slow / frequently breaks from mainland China. We provide cloud-drive mirrors (Aliyun Drive + Quark, same data, MD5-verified). Two-step setup:

:: 1. Download zip from cloud drive (link below), extract to e.g. D:\fa-data
:: 2. Wire it into your workspace:
fa data link --src D:\fa-data

fa data link writes config/loaders.yaml to point at your extracted directory — no copy, no symlink. Full walkthrough: docs/setup/data_offline.md.

Auto-acceleration since v1.0.6: fa init defaults HF_ENDPOINT=https://hf-mirror.com + enables hf_transfer multi-connection downloads (3-10× speedup) — no flag needed. Override either by setting your own env var. Power users outside CN: FA_DATA_SOURCE=hf fa init forces canonical hf.co.

Native ModelScope (魔搭) path: if the maintainer has mirrored data there (check HF_PACKAGES.*.modelscope_id), use FA_DATA_SOURCE=modelscope + pip install 'financial-analyst[modelscope]' for full-speed CN CDN downloads.

Some sub-agents and buddy tools fetch live data from sites that need a browser session or scraping bridge — OpenCLI is that bridge. It's a Node.js CLI: npm install -g @jackwener/opencli. Optional but recommended.

# Bare minimum (Node ≥ 21 prerequisite from nodejs.org)
npm install -g @jackwener/opencli
opencli --version              # verify

# THS-extra plugin (F10 / fund-flow / iwencai). Either path:
opencli plugin install https://github.com/jesson-hh/financial-analyst.git#main:opencli-plugin-ths-extra  # for pip-installed users
opencli plugin install file:///path/to/repo/opencli-plugin-ths-extra                                     # for source clones

# Chrome extension for cookie-mode collectors (xueqiu)
# https://chromewebstore.google.com/detail/opencli/ildkmabpimmkaediidaifkhjpohdnifk

# First test
fa news-collect                # default sources, ~200 items
fa doctor                      # verify all bridges OK

Step-by-step zh guide: beginner_zh.md Step 8. Xueqiu cookie-mode setup: xueqiu_setup.md.

financial-analyst is a tool-heavy 24-agent system — Tier-1 calls buddy tools, Tier-2 joins cross-stock data, Tier-3 writes structured reports with [V#]/[F#] anchors, and report-writer is the only agent allowed to touch disk. Your LLM choice decides whether the swarm uses its tools or fabricates answers from training data.

Set one provider's *_API_KEY in .env (the fa init wizard prompts for it). Defaults are loaded from config/llm.yaml.

Default ships with qwen3.5-plus. Aliyun Bailian gives 1M free token credit on signup — roughly 150 stock-deep-dive reports before you pay anything.

network_profile decides how each provider connects through Chinese network conditions (Clash fake-ip, MITM, etc.):

> /model deepseek-reasoner    # in TUI, no restart, in-flight session preserved
> /model qwen3-coder-plus     # bare name resolves provider; or use provider/model form
> /model                      # list configured models

Or change the default_provider / default_model in config/llm.yaml. See docs/llm_routing.md for the multi-provider AsyncOpenAI client design.

Personal project, single maintainer. File issues at github.com/jesson-hh/financial-analyst/issues.

• VERSIONING.md — N-2 LTS, semver policy
• docs/journey.md — bilingual build journey (empty repo → 440 alphas + 24 agents, ~2 weeks)

Apache 2.0. Research and educational purposes only. Drafts analyst-grade work product for review by qualified professionals. Does not make investment recommendations, execute transactions, or post to any ledger. You are responsible for compliance with applicable laws.

_{v1.0.7 · 2026-05-26 · made by @jesson-hh · bilingual zh/en}