Skip to main content

Documentation Index

Fetch the complete documentation index at: https://condor.hummingbot.org/llms.txt

Use this file to discover all available pages before exploring further.

Now let’s create your first Trading Agentβ€”an autonomous system that observes markets, reasons about strategy, and executes trades.

Start Agent Builder

  1. Send /agent in Telegram
  2. Select Agent Builder mode
  3. Follow the guided flow

Phase 1: Strategy Design

Tell the Agent Builder what you want to trade:
You: I want to create a scalping strategy for SOL-USDT on Binance perpetual.
     Use support/resistance levels with EMAs for trend confirmation.
     Take profit at 2%, stop loss at 1%.

Agent Builder: Great! I'll create a SOL perp scalper with:
- Market: Binance Perpetual SOL-USDT
- Entry: Support levels + EMA alignment
- Exit: 2% TP / 1% SL
- Position size: $25 per trade

Shall I proceed with strategy design?

Phase 2: Market Data

The builder selects or creates routines for market analysis:
Agent Builder: I'll create a routine to analyze SOL-USDT with:
- 1-minute candles
- EMA(7), EMA(25), EMA(99) for trend
- Support/resistance detection

Running analysis now...

Analysis Result:
- Price: $156.20
- Trend: Bullish (EMAs aligned)
- Support: $154.50, $152.00
- Resistance: $158.00, $160.50

Does this look correct?

Phase 3: Strategy Logic

The builder creates your agent.md file:
---
name: SOL Scalper
tick_interval: 60
connectors:
  - binance_perpetual

configs:
  trading_pair: SOL-USDT
  position_size: 25

limits:
  max_position_size_quote: 100
  max_single_order_quote: 25
  max_open_executors: 3
  max_drawdown_pct: 5
---

## Goal
Scalp SOL-USDT on Binance perpetual using support/resistance with EMA confirmation.

## Entry Rules
1. Run technical analysis routine
2. If trend is bullish AND price near support β†’ Long
3. If trend is bearish AND price near resistance β†’ Short

## Exit Rules
- Take profit: 2%
- Stop loss: 1%
- Keep position on stop (to recover)

## Constraints
- Max 3 concurrent positions
- Never exceed $100 total exposure

Phase 4: Dry Run

Test the agent’s reasoning without real trades:
Agent Builder: Starting dry run...

Dry Run Result:
- Analyzed SOL-USDT
- Trend: Bullish
- Price near support ($154.50)
- Decision: Would open LONG
- Reasoning: "EMAs aligned bullish, price 1.1% from S1"

No trades executed (dry run mode).
Ready for live test?

Phase 5: Run Once

Test with a single real trade:
Agent Builder: Running once with live trading...

Run Once Result:
- Created Position Executor
- Side: LONG
- Entry: $155.80
- Take Profit: $158.92 (2%)
- Stop Loss: $154.24 (1%)

Position is now active. Monitor with /agent.

Phase 6: Deploy Live

Start the agent in loop mode:
You: Deploy live with 60 second frequency

Agent Builder: Agent deployed! πŸš€

SOL Scalper is now running:
- Frequency: Every 60 seconds
- Mode: Loop
- Risk limits active

Monitor with /agent β†’ SOL Scalper β†’ View Status

Monitor Your Agent

Once deployed, monitor via Telegram:
/agent β†’ SOL Scalper β†’ View Status

πŸ€– SOL Scalper
Status: Running
Session: session_1
Tick: 47

Active Executors:
- Position #1: LONG 10 SOL @ $155.80 (+$4.20)

Recent Decisions:
- Tick 47: Monitoring, no action
- Tick 46: Opened long position
- Tick 45: Waiting for setup

Stop or Pause

/agent β†’ SOL Scalper β†’ Stop Agent
The agent stops cleanly, keeping any open positions for you to manage manually or resume later.

Next Step

Managing Agent Sessions

Learn to monitor, debug, and control agents