traderai
Self-hosted trading analysis platform pairing Claude briefings with Jev probabilities for signals like thesis-breaking headlines.
Overview
TraderAI is a self-hosted analysis platform for an individual's own portfolio. It combines a scoring engine, a risk engine enforcing the user's own rules, a discipline guardian, a backtester and a React dashboard, drawing on free data such as yfinance prices, SEC filings, Fed data and options chains. It never executes trades and is explicitly not investment advice.
It uses two models: Claude (Haiku or Sonnet) writes reasoning and morning briefings through tools that only speak from fetched data, while Jev returns calibrated probabilities for signals inside the pipeline, such as whether a headline breaks a thesis or how strongly a news item matches a tracked macro theme. The README reports total AI costs of about $1 to $2 a month. It suits technically inclined retail investors who want their data to stay local.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Yes / no | Is this headline thesis-breaking for this position? | yes · no |
| Score | How strongly does today's news match this tracked macro theme?(inferred) | — |
How it works
- Fetch prices, filings, macro data and headlines each morning.
- Jev scores headlines against positions and tracked themes.
- The scoring and risk engines check rules; Claude writes a grounded briefing.
- The user reviews and acts manually at their broker.
Strengths
- Self-hosted; portfolio data stays local.
- Analyst answers only from fetched data.
- The README reports about $1 to $2 a month in API costs.
Caveats
- Not investment advice and does not execute trades.
- Relies on hosted Claude and Jev APIs.
At a glance
- Type
- Service
- Status
- beta
- Runtime
- self-hosted service + React dashboard
- Has code
- yes
Use cases and integrations
How curated lists describe it
Self-hosted AI trading intelligence platform - scoring engine, two-model AI analyst (Claude + TypeSafe Jev), risk engine, discipline guardian, backtester, React dashboard.
hellogumbo/awesome-jev
Frequently asked questions
What does Jev do in TraderAI?
It returns calibrated probabilities for pipeline signals, such as whether a headline is thesis-breaking or matches a macro theme.
Does it trade for me?
No, it only analyzes; you confirm every action at your broker.
What does it cost to run?
The README reports about $1 to $2 a month in API costs.
Related projects
Chinese open-source research and trading workbench whose JEV monitor watches futures contracts and judges entries and exits.
ProjectbetaTypeScriptStock terminal that gathers live evidence via Valyu and asks Jev for the probability a stock trades higher in 30 days.
ProjectprototypeTypeScriptJev demo that reads live yfinance profiles and headlines into ranked, grounded and routed trade ideas, showing every request.
AppprototypePython- QuantDinger★ 12,754
Self-hosted open-source AI trading OS with research, backtesting and paper/live execution, advertising Jev-powered pre-trade decisions.
LibraryproductionPython Watches X for updates relevant to your YAML profile, has Jev classify and score each post, and alerts in Slack.
ProjectbetaTypeScriptWorking paper placing a text judge beside a live trading system as a read-only signal, comparing Jev and Claude as blind headline judges.
Paperresearch