> ## Documentation Index
> Fetch the complete documentation index at: https://docs.axlprotocol.org/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart

> Compress your first text in 3 lines. Run your first multi-LLM deliberation in 5 minutes.

## Install

```bash theme={null}
pip install axl-core
```

## Compress Text

```python theme={null}
from axl import compress

text = "The Federal Reserve held rates at 5.25%. Markets dropped 0.3%."
packets = compress(text)
print(packets)
```

```
@axlprotocol.org/rosetta
ID:FED-01|OBS.99|!rate_hold|@fed.powell|$rate:5.25%|NOW
ID:MKT-01|OBS.95|$SPY|<-!rate_hold|^-0.3%|NOW
```

## Parse a Packet

```python theme={null}
from axl import parse

pkt = parse("ID:WHL-01|OBS.95|!whale_move|@whale.CW2024|^2400BTC=>exchange|NOW")
print(pkt.operation)   # OBS
print(pkt.confidence)  # 0.95
print(pkt.subject)     # !whale_move
```

## Run the AXL Silo

```bash theme={null}
git clone https://github.com/axlprotocol/axl-silo.git
cd axl-silo && pip install -r requirements.txt
python run.py --port 7000
```

Open `http://localhost:7000`. Mix providers (Anthropic + OpenAI + Google + Local). Pick a seed. Hit ENGAGE. Watch LLMs debate in AXL.