Jupiter Signal

AI-assisted correlation engine for Jupiter's Developer Platform.

jupiter-signal combines Jupiter Price, Tokens and Prediction Markets APIs to produce structured token intelligence for AI agents and analyst workflows. It is designed as an Atlas Nexus proof-of-work project: API composition, clean CLI output, JSON-first automation and readable reasoning.

Built for Superteam Earn's Not Your Regular Bounty / Frontier Hackathon track.

Live mini demo

A static project page is available here:

https://atlasnexusops.github.io/jupiter-signal/

What it does

Price API ─────┐
               ├──→ Correlation Engine ──→ JSON / CLI report
Tokens API ────┤        │
               │        └──→ Agent-readable reasoning
Prediction ────┘
Markets API

The engine cross-references independent Jupiter data surfaces:

1. Price API — token price and short-term change.
2. Tokens API — verification, organic score and market metadata.
3. Prediction Markets API — related crypto events and sentiment context.
4. Correlation engine — flags divergence between spot momentum, token quality and prediction-market context.
5. Structured output — emits compact JSON for downstream agents, dashboards or review workflows.

Example output

{
  "token": {
    "symbol": "JUP",
    "name": "Jupiter"
  },
  "price": {
    "usd": 0.4056,
    "change_24h_pct": 52.9
  },
  "quality": {
    "organic_score": 92,
    "verified": true
  },
  "prediction_context": {
    "related_markets": 3,
    "sentiment": "bullish"
  },
  "signal": {
    "type": "momentum_divergence",
    "confidence": 0.85,
    "reasoning": "+52.9% 24h price move, organic score 92/100, verified token, related prediction markets leaning bullish. Review manually before acting."
  }
}

CLI usage

pip install -e .
export JUPITER_API_KEY="your-key"

jupiter-signal scan SOL JUP BONK
jupiter-signal scan SOL JUP --json
jupiter-signal markets crypto

Why it matters

Jupiter's Developer Platform exposes several useful APIs. The value of this project is in combining them rather than treating them as isolated endpoints.

Useful patterns demonstrated:

• multi-API correlation;
• async REST data collection;
• JSON-first output for AI agents;
• analyst-readable reasoning;
• compact CLI workflow;
• bounty-grade documentation and DX notes.

Repository structure

jupiter_signal/
├── __init__.py      # Package metadata
├── apis.py          # Jupiter API wrappers and correlation engine
└── cli.py           # CLI interface with Rich terminal output + JSON export

docs/
└── index.html       # Static GitHub Pages mini demo

Built with Jupiter's AI stack

During development, the project used Jupiter's agent-facing documentation surfaces, including Skills and llms.txt. Notes are captured in DX-REPORT.md.

Scope and disclaimer

This project is a technical demonstration and research workflow. It is not financial advice and does not execute transactions. Any market interpretation should be reviewed manually before operational use.

License

MIT — Atlas Nexus, 2026