jev-trading
Local stock decision service that sends market data and news to Jev or a local model and returns a policy-checked buy/sell/hold.
Overview
Jev Trading is a stock decision service that runs on your computer, with a web workbench and an HTTP API. You submit a symbol, current position and horizon (for example a Chinese A-share code, flat, next five trading days); it gathers market data, indicators, fundamentals and news collected by a companion tool called AIStock, asks a model for probabilities over buy, sell and hold, and then applies policy checks that can change the suggestion, for example turning a buy into a hold when technical data is degraded.
Results include the model's original action, the final action, a status (accepted, blocked, skipped, error or expired), reason codes, an expiry time and action probabilities, which the README stresses are not profit odds. Inference can use Jev in the cloud or a local model through the same API, and history, input snapshots and raw responses are kept locally with export. It is signals only: no brokerage integration, stop-losses, sizing or return backtests. A demo mode runs without keys. Documentation is available in five languages.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Given this stock's data, position and horizon, should the action be buy, sell or hold? | buy · sell · hold |
How it works
- Submit symbol, position and horizon via UI or HTTP
- Collect market data, indicators, fundamentals and news
- Model returns action probabilities
- Policy checks set the final action, status, reasons and expiry
Strengths
- Traceable evidence and raw responses stored locally
- Policy layer can override the model on data-quality issues
- Cloud Jev or local model through one API
- Demo mode without keys
Caveats
- Signals only; no orders or backtests
- Probabilities are not profit odds
- Validated on macOS only
Getting started
Install Bun and uv, start the service, then `curl -sS http://127.0.0.1:3000/v1/decisions -H 'Content-Type: application/json' -d '{"mode":"demo","position":"flat"}'`.At a glance
- Type
- Project
- Status
- beta
- Language
- TypeScript
- Runtime
- Bun service + Python
- Has code
- yes
Use cases and integrations
How curated lists describe it
\*\*Jev Trading is a stock decision service that runs on your computer.\*\* Use the web workbench or call it from your own software over HTTP. It sends market data, indicators, fundamentals, and news collected by AIStock to a model, then returns a structured decision with a record of the evidence.
logicrw/awesome-jev-projects
Frequently asked questions
Does it place trades?
No. It produces signals only, with no brokerage integration.
Can the final action differ from the model's?
Yes, policy checks can override it, for example when data quality is degraded.
Can I use a local model?
Yes, cloud Jev and local models share the same API.
Related projects
Local China A-share technical analysis workbench where Jev assesses eight indicator dimensions and returns Buy, Hold or Sell with evidence.
ProjectbetaPython, TypeScriptPyTorch factor-research stack with 213 factors and cost-aware backtests, plus a preview typed portfolio-decision adapter.
ProjectbetaPythonChinese open-source research and trading workbench whose JEV monitor watches futures contracts and judges entries and exits.
ProjectbetaTypeScriptSelf-hosted trading analysis platform pairing Claude briefings with Jev probabilities for signals like thesis-breaking headlines.
Servicebeta- Jev-Trades★ 51
Next.js crypto dashboard where Jev judges each tick across timeframes and drives a SQLite paper-trading portfolio with automatic TP/SL.
ServiceprototypeTypeScript Educational multi-agent AI hedge fund that paper-trades, with TypeSafe Jev as one supported model provider for its investor agents.
ProjectprototypePython