coopersimson96

Kronos Claude Skill — AI skill for Claude Code

AI community

Wall Street's billion dollar stock prediction AI as a Claude skill.

How to install Kronos Claude Skill

This entry records only its repository, not the path inside it, so there is no exact command to give. Open coopersimson96/kronos-claude-skill and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Kronos Claude Skill does

Wall Street's billion dollar stock prediction AI as a Claude skill. Run hedge-fund-grade technical analysis from any Claude conversation in plain English.

Alternatives in AI

AIHawk — Open-source AI browser agent: describe any task in plain language and it autonomously browses, clicks, types a 30.3k ★
CLI 代理 API — English 中文一个为 CLI 提供 OpenAI/Gemini/Claude/Codex 兼容 API 接口的代理服务器 19k ★
Mirofish CLI — Multi-agent AI prediction engine - digital sandbox for scenario simulation (fork of 666ghj/MiroFish) 307 ★

README

Kronos — Claude Skill

The same kind of price prediction AI Wall Street pays millions for, packaged as a Claude skill you can invoke in plain English.

**Not financial advice.** This is a research tool. Kronos outputs a probability distribution with a confidence interval, not a price target. Use it as one input into a research process. Stocks carry real risk.

What it does

Run hedge-fund-grade technical analysis on any stock or crypto ticker from any Claude conversation.

You: run kronos on AAPL
Claude: [invokes the skill, returns prediction + confidence interval + interpretation]

Under the hood: the skill pulls recent OHLCV via yfinance, runs the Kronos foundation model (trained on 12B candlesticks from 45 exchanges) for a forward prediction, and returns formatted markdown Claude reads back to you in plain English.

Install

One-time. First run takes 3-7 minutes to clone the Kronos repo, create a venv, and install torch + transformers + yfinance.

# The skill auto-installs on first invocation. You can also pre-install:
python3 ~/.claude/skills/kronos/scripts/run_kronos.py AAPL

Usage

From any Claude conversation:

run kronos on AAPL — 6 months of daily candles, 24-day forecast (defaults)
kronos BTC-USD 3mo 4h 48 — 3 months of 4-hour candles, 48-period forecast
predict NVDA with kronos for the next 30 days — claude infers args

Or invoke the script directly:

python3 ~/.claude/skills/kronos/scripts/run_kronos.py  [period] [interval] [pred_len]

What the output looks like

# Kronos Prediction: AAPL

**Last close:** $XXX.XX (YYYY-MM-DD)
**Predicted close after 24 x 1d candles:** $XXX.XX
**Direction:** UP (+X.XX%)
**Confidence band:** NARROW (signal worth investigating)
**Forecast range:** $XXX.XX - $XXX.XX (±X.XX% of last close)

## Interpretation rules
- NARROW band (<5%): model is confident...
- MODERATE band (5-10%): mixed signal...
- WIDE band (>10%): treat as noise...