Question 1

Is the trade signal actually Jev, or hand-tuned indicators wearing a Jev label?

Accepted Answer

Both, and the README is explicit about the split. Python first computes engagement velocity, author diversity ratio, and keyword sentiment polarity, then Jev answers four typed questions over that combined state: trade action, sentiment spectrum, squeeze risk, and catalyst significance.

Question 2

Is this financial advice, and can it place trades for me?

Accepted Answer

It never executes trades. You get a decision card with an action, a calibrated confidence percentage, an entry range, a stop loss, and target levels that you place manually on your own exchange or DEX. The README states outright that none of its output constitutes financial or trading advice.

Question 3

What does a single search cost me?

Accepted Answer

TwitterAPI.io dominates: roughly $0.0075 for 50 tweets up to about $0.15 for 1,000, while the Jev pass is around $0.0008 either way. A repeat search for the same asset inside the 10-minute window costs nothing, because pagination halts at the first tweet ID already in the local SQLite store.

Question 4

Will it expose the app or my keys beyond localhost?

Accepted Answer

It binds to 127.0.0.1 by default with configurable HOST and PORT, and ships ALLOWED_ORIGINS plus an optional ADMIN_TOKEN for access control. Keys are read from a .env file you create. Beyond those knobs the README does not document authentication on the API routes.

Question 5

How do I install and run Jev X Sentiment Analysis?

Accepted Answer

Requires Python 3.12 or higher, a TypeSafe API key and a TwitterAPI.io key. Create a venv, `pip install -r requirements.txt`, add TYPESAFE_API_KEY and TWITTER_API_KEY to .env, then run `uvicorn app.main:app --reload --port 8000` and open http://localhost:8000.

Question 6

What four questions does Jev answer?

Accepted Answer

It evaluates four typed questions concurrently on the combined state: Trade Action as a Choice from Strong Buy to Strong Sell, Sentiment Spectrum as a Score from extreme panic to euphoria, Squeeze Risk as a Noul, and Catalyst Significance as a Score from none to major.

Question 7

How does it keep repeat-search API costs down?

Accepted Answer

Tweets are stored in a local SQLite database and returned in reverse-chronological order, so pagination halts as soon as a returned tweet ID already exists and the rest of the sample loads from the database. Repeated searches for the same asset within a 10-minute window hit the cache and cost $0.00.

Jev X Sentiment Analysis

About this project

Highlights

Quickstart

Questions about this entry

Reactions & coverage

From the Jev guides

From the Jev AI model community