Colosseum Agent Hackathon 2026

The first AI agent that proves it's getting smarter

Self-learning AI with on-chain proof of growth. Watch it evolve, measure its progress, verify on Solana.

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                    Cortex AGENT                           β”‚
β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€
β”‚  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”   β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”   β”Œβ”€β”€β”€β”€β”€β”                    β”‚
β”‚  β”‚PERCEIVE │──▢│ REASON │──▢│ ACT β”‚                    β”‚
β”‚  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜   β””β”€β”€β”€β”€β”€β”€β”€β”€β”˜   β””β”€β”€β”€β”€β”€β”˜                    β”‚
β”‚       β–²                         β”‚                       β”‚
β”‚       β”‚    β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜                       β”‚
β”‚       β”‚    β–Ό                                            β”‚
β”‚       β”‚  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”   β”‚
β”‚       β”‚  β”‚           REFLECT & LEARN               β”‚   β”‚
β”‚       β”‚  β”‚  β€’ Score outcome (reward signal)        β”‚   β”‚
β”‚       β”‚  β”‚  β€’ Update strategy effectiveness        β”‚   β”‚
β”‚       β”‚  β”‚  β€’ Generate new strategy variants       β”‚   β”‚
β”‚       β”‚  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜   β”‚
β”‚       β”‚                   β”‚                             β”‚
β”‚       β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜                             β”‚
β”‚                           β–Ό                             β”‚
β”‚              β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”                   β”‚
β”‚              β”‚  SOLANA MILESTONE   β”‚ ← On-chain proof  β”‚
β”‚              β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜                   β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
        

How Cortex Learns

πŸ‘οΈ

1. Perceive

Gather observations from environment β€” prices, news, chain data, web searches.

πŸ€”

2. Reason

Select best strategy based on past success rates. Balance exploration vs exploitation.

⚑

3. Act

Execute planned actions using available tools. Handle errors, try fallbacks.

πŸ”

4. Reflect

Score outcomes. What worked? What failed? Generate improvement suggestions.

πŸ“ˆ

5. Learn

Update strategy success rates. Create new strategy variants from failures.

⛓️

6. Prove

Record significant improvements as milestones on Solana. Verifiable growth.

Live Demo

Watch Cortex learn in real-time with Claude LLM

πŸ“Š Learning Metrics

0
Actions
0%
Success Rate
0%
Avg Reflection
0
Strategies
0
⛓️ Milestones

Success Rate Over Time

πŸ”„ Agent Activity

Click "Start Demo" to begin...

Available Tools

Real APIs that Cortex can use (x402 micropayments)

πŸ”

Web Search

Brave API integration

$0.002/call
πŸ“„

URL Fetch

Content extraction

$0.001/call
🌀️

Weather

Open-Meteo API

$0.001/call
πŸ“ˆ

Crypto Prices

CoinGecko API

$0.001/call
πŸ‘›

Wallet Analysis

Solana RPC / Helius

$0.003/call
πŸ“°

News

Google News RSS

$0.001/call
🎨

Image Gen

DALL-E 3

$0.02/call

Ready to build self-learning agents?

Cortex is open source. Start building today.