Browse Robonet Data

Browse Robonet Data

Quick Start

This skill provides fast, read-only access to explore Robonet's trading resources before building anything. All tools execute in under 1 second and cost little to nothing.

Load the tools first:

Use MCPSearch to select: mcp__workbench__get_all_symbols
Use MCPSearch to select: mcp__workbench__get_all_technical_indicators
Use MCPSearch to select: mcp__workbench__get_data_availability

Common starting pattern:

1. get_all_symbols → See available trading pairs (BTC-USDT, ETH-USDT, etc.)
2. get_all_technical_indicators → Browse 170+ indicators (RSI, MACD, Bollinger Bands)
3. get_data_availability → Check data ranges before backtesting

When to use this skill:

• Start every workflow by exploring available resources
• Check data availability before building strategies
• Review existing strategies and their performance
• Understand what ML predictions are available (Allora topics)
• Audit recent backtest results

Available Tools (8)

Strategy Exploration Tools

get_all_strategies - List your trading strategies with optional backtest results

• Parameters:
  • include_latest_backtest (optional, boolean): Include latest backtest summaries
• Returns: List of strategies with names, components, and optionally performance metrics
• Pricing: $0.001
• Use when: Finding existing strategies to review, enhance, or compare

get_strategy_code - View Python source code of a strategy

• Parameters:
  • strategy_name (required, string): Name of the strategy
• Returns: Complete Python source code
• Pricing: Free
• Use when: Learning from existing strategies, reviewing before modification, debugging

get_strategy_versions - Track strategy evolution across versions

• Parameters:
  • base_strategy_name (required, string): Base name without version suffixes
• Returns: List of all versions with creation dates and modification history
• Pricing: $0.001
• Use when: Understanding how a strategy evolved, comparing versions, auditing changes

Market Data Tools

get_all_symbols - List tradeable pairs on Hyperliquid Perpetual

• Parameters:
  • exchange (optional, string): Filter by exchange name
  • active_only (optional, boolean): Only return active symbols (default: true)
• Returns: List of symbols with exchange, symbol name, active status, backfill status
• Pricing: $0.001
• Use when: Choosing which assets to trade, checking what's available before building strategies

get_data_availability - Check data ranges before backtesting

• Parameters:
  • data_type (optional, string): Type of data (crypto, polymarket, all)
  • symbols (optional, array): Specific crypto symbols to check
  • exchange (optional, string): Filter crypto by exchange
  • asset (optional, string): Filter Polymarket by asset
  • include_resolved (optional, boolean): Include resolved Polymarket markets
  • only_with_data (optional, boolean): Only show items with available data
• Returns: Data availability with date ranges, candle counts, backfill status
• Pricing: $0.001
• Use when: Before backtesting (verify sufficient data), choosing test date ranges, checking market coverage

Indicator & ML Tools

get_all_technical_indicators - Browse 170+ indicators available in Jesse framework

• Parameters:
  • category (optional, string): Filter by category (momentum, trend, volatility, volume, overlap, oscillators, cycle, all)
• Returns: List of indicators with names, categories, and parameters
• Pricing: $0.001
• Use when: Exploring indicators for strategy ideas, checking parameter requirements, learning what's available

get_allora_topics - List Allora Network ML prediction topics

• Parameters: None
• Returns: List of topics with asset names, network IDs, prediction horizons, and prediction types
• Pricing: $0.001
• Use when: Planning ML enhancement, checking prediction coverage, understanding available horizons (5m, 8h, 24h, 1 week)

Backtest Results Tool

get_latest_backtest_results - View recent backtest performance

• Parameters:
  • strategy_name (optional, string): Filter by strategy name
  • limit (optional, integer, 1-100): Number of results (default: 10)
  • include_equity_curve (optional, boolean): Include equity curve timeseries
  • equity_curve_max_points (optional, integer, 50-1000): Maximum points for equity curve
• Returns: List of backtest records with metrics (Sharpe, drawdown, win rate, total return, profit factor)
• Pricing: Free
• Use when: Checking if backtest already exists, comparing strategy performance, avoiding redundant backtests

Core Concepts

Symbol Coverage

Crypto Perpetuals (Hyperliquid):

• Major pairs: BTC-USDT, ETH-USDT, SOL-USDT, NEAR-USDT
• Data history: BTC-USDT and ETH-USDT have longest history (2020-present)
• Typical range: Most symbols have 6-24 months of data
• Data quality: 1-minute candles available for all symbols

Best practices:

• Use get_all_symbols to see complete list
• Check get_data_availability for specific symbol history
• BTC-USDT and ETH-USDT recommended for initial strategy development (longest history)

Technical Indicators

170+ indicators organized by category:

• Momentum (16 indicators): RSI, MACD, Stochastic, ADX, CCI, MFI, ROC, Williams %R, Ultimate Oscillator, etc.
• Trend (12 indicators): EMA, SMA, DEMA, TEMA, WMA, Supertrend, Parabolic SAR, VWAP, HMA, etc.
• Volatility (8 indicators): Bollinger Bands, ATR, Keltner Channels, Donchian Channels, Standard Deviation, etc.
• Volume (10 indicators): OBV, Volume Profile, Chaikin Money Flow, Volume Weighted indicators, etc.
• Overlap (8 indicators): Various moving averages and envelopes
• Oscillators (6 indicators): Specialized momentum oscillators
• Cycle (4 indicators): Market cycle detection indicators

How to use:

1. get_all_technical_indicators(category="momentum") → Browse momentum indicators
2. Pick indicators for your strategy concept
3. Reference indicators in strategy description when building

Note: All indicators are from the Jesse framework (jesse.indicators). Use exact names when creating strategies.

Allora Network ML Predictions

Prediction Coverage:

• Assets: BTC, ETH, SOL, NEAR
• Horizons: 5 minutes, 8 hours, 24 hours, 1 week
• Prediction types:
  • Log return (percentage change prediction)
  • Absolute price (future price prediction)
• Networks:
  • Mainnet: 10 production topics
  • Testnet: 26 experimental topics

Topic structure:

Asset: BTC
Horizon: 24h
Type: Log return
Network: mainnet

How to use:

1. get_allora_topics() → See all available predictions
2. Match prediction horizon to your strategy timeframe
3. Use enhance_with_allora (from improve-trading-strategies skill) to integrate predictions

Best practices:

• Match prediction horizon to strategy timeframe (don't use 5m predictions for daily strategy)
• Mainnet topics are production-ready, testnet topics are experimental
• Check topic availability before planning ML enhancement

Backtest Result Interpretation

Key Metrics:

Sharpe Ratio (risk-adjusted return):

• >2.0: Excellent performance
• 1.0-2.0: Good performance
• 0.5-1.0: Acceptable performance
• <0.5: Poor performance

Max Drawdown (largest peak-to-trough decline):

• <10%: Conservative risk profile
• 10-20%: Moderate risk profile
• 20-40%: Aggressive risk profile
• >40%: Very risky (reconsider strategy)

Win Rate (percentage of profitable trades):

• 45-65%: Realistic for most strategies
• >70%: Suspicious (possible overfitting or unrealistic fills)
• <40%: Needs improvement

Profit Factor (gross profit / gross loss):

• >2.0: Excellent
• 1.5-2.0: Good
• 1.2-1.5: Acceptable
• <1.2: Marginal (risky to deploy)

How to use backtest results:

1. get_latest_backtest_results(strategy_name="MyStrategy") → Check recent tests
2. Review metrics against benchmarks above
3. If metrics good: consider deployment
4. If metrics poor: refine strategy or try different approach

Best Practices

Exploration Workflow

Start every strategy development with data exploration:

1. Explore available assets
   get_all_symbols() → What can I trade?
   get_data_availability(data_type="crypto") → How much history?

2. Understand available tools
   get_all_technical_indicators(category="momentum") → What indicators?
   get_allora_topics() → What ML predictions available?

3. Review existing work
   get_all_strategies(include_latest_backtest=true) → What's already built?
   get_strategy_code(strategy_name="Existing") → Learn from existing code

4. Plan your strategy
   → Use insights from exploration to inform strategy design

Data Availability Checks

Always verify sufficient data before backtesting:

Problem: Backtest fails with "No data available"
Solution:
  1. get_data_availability(symbols=["BTC-USDT"],
  2. Check date range returned
  3. Use date range within available data for backtest

Minimum data requirements:

• Quick test: 1-3 months (limited validation)
• Standard test: 6-12 months (recommended minimum)
• Robust test: 12-24 months (ideal for validation)

Cost Optimization

All tools in this skill are cheap (free to $0.001):

• Use liberally during exploration
• No need to batch queries or optimize calls
• Better to over-explore than under-explore

Cost-saving pattern:

1. Browse data (this skill, <$0.01) → Explore thoroughly
2. Generate ideas (design-trading-strategies, $0.05-$1.00) → Cheap exploration
3. Build strategy (build-trading-strategies, $1-$4.50) → Expensive, be sure first

Spending 2-3 minutes exploring data (costs <$0.01) can save dollars in wasted strategy generation.

Learning from Existing Strategies

Use existing strategies as templates:

1. get_all_strategies(include_latest_backtest=true)
   → Find high-performing strategies (Sharpe >1.5)

2. get_strategy_code(strategy_name="HighPerformer")
   → Study the code structure

3. Identify patterns:
   - How are entry conditions structured?
   - What indicators are used?
   - How is position sizing calculated?
   - How is risk management implemented?

4. Apply learnings to new strategy design

Indicator Research

Find the right indicators for your strategy concept:

Strategy Type → Indicator Categories to explore:
- Trend Following → trend, momentum
- Mean Reversion → oscillators, momentum
- Breakout → volatility, volume
- Scalping → momentum, volume
- Swing Trading → trend, overlap

Example exploration:

Building a mean reversion strategy:
1. get_all_technical_indicators(category="oscillators") → See oscillators
2. get_all_technical_indicators(category="momentum") → See momentum indicators
3. Pick RSI (overbought/oversold) + Bollinger Bands (deviation from mean)
4. Use these indicator names when building strategy

Common Workflows

Workflow 1: Pre-Strategy Exploration

Goal: Understand what's available before building anything

1. get_all_symbols()
   → Review available trading pairs
   → Note which symbols interest you

2. get_data_availability(symbols=["BTC-USDT", "ETH-USDT"],
   → Check data ranges for chosen symbols
   → Verify sufficient history (6+ months preferred)

3. get_all_technical_indicators(category="all")
   → Browse all 170+ indicators
   → Note which indicators fit your strategy idea

4. get_allora_topics()
   → See ML prediction coverage
   → Check if your asset has predictions available
   → Note prediction horizons

5. Ready to build:
   → If exploring ideas: Use design-trading-strategies skill
   → If ready to code: Use build-trading-strategies skill

Cost: ~$0.005 (essentially free)

Workflow 2: Strategy Audit

Goal: Review existing strategies and their performance

1. get_all_strategies(include_latest_backtest=true)
   → See all strategies with performance data

2. Identify interesting strategies:
   → High Sharpe ratio (>1.5)
   → Acceptable drawdown (<20%)
   → Realistic win rate (45-65%)

3. get_strategy_code(strategy_name="TopPerformer")
   → Review implementation details
   → Understand why it performs well

4. get_strategy_versions(base_strategy_name="TopPerformer")
   → See how strategy evolved
   → Identify what improvements were made

5. Apply learnings:
   → Use as template for new strategies
   → Or enhance further with improve-trading-strategies skill

Cost: Free to $0.003

Workflow 3: Data Coverage Check

Goal: Verify data availability before backtesting

1. Choose your strategy parameters:
   Symbol: BTC-USDT
   Timeframe: 1h
   Test period: 6 months

2. get_data_availability(symbols=["BTC-USDT"],
   Returns: "BTC-USDT available from 2020-01-01 to 2025-02-02"

3. Verify coverage:
   ✓ Has 6+ months of data
   ✓ Covers desired test period
   ✓ Ready to backtest

4. If insufficient data:
   → Try shorter test period
   → Or choose different symbol (BTC-USDT and ETH-USDT have longest history)

5. Proceed to testing:
   → Use test-trading-strategies skill to run backtest

Cost: $0.001

Workflow 4: Indicator Research

Goal: Find the right indicators for your strategy concept

Strategy Concept: Mean reversion on cryptocurrency

1. get_all_technical_indicators(category="momentum")
   → Browse momentum indicators (RSI, Stochastic, etc.)

2. get_all_technical_indicators(category="volatility")
   → Browse volatility indicators (Bollinger Bands, ATR, etc.)

3. Select indicators for mean reversion:
   → RSI (identify overbought/oversold)
   → Bollinger Bands (measure deviation from mean)
   → ATR (position sizing based on volatility)

4. Note exact indicator names:
   → "RSI" (not "rsi" or "RelativeStrengthIndex")
   → "BollingerBands" (not "BB" or "bollinger")
   → "ATR" (not "AverageTrueRange")

5. Use exact names in strategy description:
   → When using build-trading-strategies skill
   → Reference indicators precisely: "Use RSI with period 14"

Cost: $0.002

Advanced Usage

Filtering and Optimization

Efficient querying:

# Get only active symbols
get_all_symbols(active_only=true)

# Filter indicators by category
get_all_technical_indicators(category="momentum")

# Check specific symbols only
get_data_availability(symbols=["BTC-USDT", "ETH-USDT"],

# Limit backtest results
get_latest_backtest_results(limit=5)

Backtest Result Analysis

Detailed equity curve analysis:

get_latest_backtest_results(
    strategy_name="MyStrategy",
    include_equity_curve=true,
    equity_curve_max_points=500
)

Returns equity curve data for visualizing strategy performance over time.

Use cases:

• Identify periods of strong/weak performance
• Detect regime changes (strategy works in trending vs ranging markets)
• Compare multiple strategies visually

Cross-Asset Research

Compare data availability across assets:

1. get_data_availability(data_type="crypto",
   → See all crypto pairs with data

2. Compare:
   - Which symbols have longest history?
   - Which symbols have most recent backfills?
   - Which timeframes are well-covered?

3. Choose optimal symbols for strategy development:
   → BTC-USDT, ETH-USDT: Longest history, most reliable
   → Altcoins: Shorter history, higher risk, potentially higher returns

Troubleshooting

"No Strategies Found"

Issue: get_all_strategies returns empty list

Solutions:

• Strategies are linked to your API key's wallet
• Ensure you're using the correct API key
• If new account, you haven't created strategies yet (use build-trading-strategies skill to create first strategy)

"Symbol Not Found"

Issue: get_data_availability doesn't show expected symbol

Solutions:

• Use get_all_symbols() to see complete list of available symbols
• Check spelling (BTC-USDT not BTC-USD or BTCUSDT)
• Some symbols may not have data backfilled yet (check active_only=false to see inactive symbols)

"No Indicator Matches Description"

Issue: Can't find indicator you're looking for

Solutions:

• Use get_all_technical_indicators(category="all") to browse complete list
• Search for similar names (RSI vs RelativeStrengthIndex)
• Check category filter (momentum indicator won't show if filtering by trend)
• Jesse framework uses specific names - use exact names returned by tool

"Backtest Results Missing"

Issue: get_latest_backtest_results doesn't show expected backtest

Solutions:

• Check strategy name spelling (case-sensitive)
• Backtest may still be running (wait 20-60 seconds)
• Backtest may have failed (check for error messages)
• Use limit parameter to retrieve more results (default is 10)

Next Steps

After exploring data with this skill:

Generate strategy ideas:

• Use design-trading-strategies skill to generate AI-powered strategy concepts
• Cost: $0.05-$1.00 per idea generation (cheapest AI tool)
• Best when: You want to explore creative concepts before committing to development

Build strategies directly:

• Use build-trading-strategies skill to generate complete strategy code
• Cost: $1.00-$4.50 per strategy (most expensive AI tool)
• Best when: You already know what you want to build

Test existing strategies:

• Use test-trading-strategies skill to backtest strategies
• Cost: $0.001 per backtest
• Best when: You have strategy code and want to validate performance

Improve strategies:

• Use improve-trading-strategies skill to refine, optimize, or enhance with ML
• Cost: $0.50-$4.00 per operation
• Best when: You have an existing strategy that needs improvement

Prediction market trading:

• Use trade-prediction-markets skill for Polymarket YES/NO token strategies
• Cost: $0.001-$4.50 depending on operation
• Best when: You want to trade on real-world events (politics, economics, sports)

Summary

This skill provides fast, cheap, read-only access to Robonet's trading resources:

• 8 data tools covering strategies, symbols, indicators, ML topics, and backtest results
• <1 second execution for all tools
• Free to $0.001 cost (essentially free to explore)
• Zero risk (read-only operations, no modifications or executions)

Core principle: Explore thoroughly before building. Spending 2-3 minutes browsing data (costs <$0.01) can save dollars in wasted strategy generation and prevent costly mistakes.

Best practice: Start every workflow with this skill, then progress to design → build → improve → test → deploy based on your findings.