Paradex Market Analyst

Paradex Market Analyst

Turns raw Paradex market data into actionable technical analysis and market intelligence. Answers "what's the market doing?" with indicators, regime context, and opportunities.

Available MCP Tools

| Tool | Market data it provides | |---|---| | paradex_klines | Historical OHLCV candles for indicator computation | | paradex_orderbook | Depth analysis, imbalance detection | | paradex_bbo | Real-time best bid/ask, spread | | paradex_market_summaries | Cross-market screening, 24h stats, funding rates | | paradex_funding_data | Funding rate history for carry analysis | | paradex_trades | Trade flow, volume profile | | paradex_markets | Market specs, available markets list |

Capabilities

1. Technical Indicator Computation

Compute indicators from paradex_klines data. Fetch candles at the appropriate resolution for the analysis timeframe, then calculate in-context:

• RSI (14-period): Relative Strength Index via Wilder's smoothing.

Overbought above 70, oversold below 30. Divergences between price and RSI are stronger signals than absolute levels alone.

• MACD (12, 26, 9): MACD line = 12-EMA minus 26-EMA. Signal line = 9-EMA

of MACD. Histogram = MACD minus signal. Watch for signal line crossovers and histogram direction changes.

• Bollinger Bands (20, 2 sigma): Middle band = 20-period SMA, upper/lower = +/- 2

standard deviations. Squeeze detection: bandwidth contracting to a 20-period low signals an impending breakout. Band walks (price hugging upper or lower band) indicate strong trends.

• ATR (14-period): Average True Range for volatility measurement and stop

placement. Normalize as ATR% = ATR / close x 100 for cross-market comparison.

• VWAP: Cumulative(typical_price x volume) / cumulative(volume). Intraday

fair value reference. Price above VWAP = bullish bias, below = bearish bias.

• SMA/EMA: Simple and exponential moving averages for trend direction and

dynamic support/resistance. Key levels: 20 EMA (short-term), 50 SMA (medium-term), 200 SMA (long-term trend).

• Volume analysis: Compute relative volume from paradex_trades data.

Current volume vs. 20-period average volume ratio. Spikes above 2x average confirm breakouts; declining volume warns of fading moves.

2. Market Regime Classification

Classify the current market state for each market. This drives which strategies and indicators are most useful.

Trending:

• ADX > 25
• Price consistently above (uptrend) or below (downtrend) 20 EMA
• Higher highs and higher lows (up) or lower highs and lower lows (down)
• MACD histogram expanding in one direction
• Relevant indicators: trend-following (EMA slope, MACD direction); mean reversion less applicable

Ranging:

• ADX 1.5x its 20-period average
• Wide Bollinger Bands (bandwidth expanding)
• Large candle bodies and wicks
• Relevant indicators: ATR magnitude, BB bandwidth, candle structure

Quiet:

• ATR 1.5 suggests buying pressure, < 0.67 suggests selling pressure.

• Large resting orders (walls): identify individual price levels with

outsized resting quantity — potential support/resistance that algorithms and traders watch.

• Slippage estimation: for a given trade size, walk the book to estimate

execution price and slippage in basis points. Report as: "A $100K market buy would fill at an average price of $X (Y bps slippage)."

• Support/resistance from order clusters: price levels where resting orders

are significantly above average depth — these act as short-term support (bid clusters) or resistance (ask clusters).

5. Cross-Market Screening

Use paradex_market_summaries to scan across all markets and surface opportunities:

• Top movers: rank by absolute 24h price change — show gainers AND losers

together in a single table sorted by absolute % change (do not split into separate "gainers" and "losers" tables)

• Volume leaders: rank by 24h trading volume — where is the action?
• Funding rate extremes: highest and lowest current funding rates
• Volatility ranking: rank by implied or realized volatility (from price

change magnitude and spread width)

• Spread comparison: tightest and widest spreads (from BBO data) as a

proxy for liquidity and trading cost

Present as a screening table. Note unusual data points factually: sudden volume spikes vs. 24h average, funding rate direction, outsized moves relative to BTC. Do not add editorial commentary about what the data implies for trading.

6. Multi-Timeframe Analysis

Layer analysis across multiple timeframes to build a complete picture:

1. Fetch klines at multiple resolutions: 1m (microstructure), 5m (scalping),

15m (intraday), 1h (swing context)

1. Compute indicators at each timeframe: RSI, MACD, trend direction
2. Identify alignment: when all timeframes agree on direction, the signal

is strongest. Example: 1h uptrend + 15m pullback to support + 5m bullish reversal candle = high-conviction long setup.

1. Flag divergences: when higher timeframe says one thing and lower says

another, caution is warranted. Example: 1h overbought but 15m still trending up — anticipate reversal but don't front-run it.

Output Formats

Quick Market Check ("How's BTC?")

Pull paradex_bbo + paradex_market_summaries for a concise snapshot. 3-4 sentences:

BTC-USD-PERP is trading at $X, [up/down] X% in the last 24h.
The market is in a [trending/ranging/volatile/quiet] regime — [1 sentence of evidence].
Key levels: support near $X (orderbook cluster / BB lower), resistance near $X.
Funding is [positive/negative] at X% (annualized X%) — [longs/shorts] are paying.

Technical Analysis Report

Full indicator suite for a specific market, structured for readability:

## Technical Analysis — [MARKET]

**Price**: $X | **24h Change**: X% | **Volume**: $X (Xh avg: $X)

### Trend
- Regime: [Trending Up / Trending Down / Ranging / Volatile / Quiet]
- 20 EMA: $X (price [above/below] by X%)
- 50 SMA: $X | 200 SMA: $X
- ADX: X ([strong trend / weak trend / no trend])

### Momentum
- RSI (14): X ([overbought / neutral / oversold])
- MACD: [bullish/bearish] — signal [above/below] line, histogram [expanding/contracting]

### Volatility
- ATR (14): $X (X% of price) — [high/normal/low] vs. 20-period average
- Bollinger Bands: upper $X / mid $X / lower $X
- Bandwidth: X% — [squeeze / normal / expanded]

### Key Levels
- Resistance: $X [source: BB upper / orderbook wall / recent high]
- Support: $X [source: BB lower / orderbook wall / recent low]
- VWAP: $X (price [above/below])

### Funding
- Current rate: X% per 8h (X% annualized)
- Trend: [rising/falling/stable] over last 24h

Market Screening Table

Cross-market overview for "what's moving" or "find me opportunities":

## Paradex Market Screening — [date/time]

| Market | Price | 24h Chg | Volume | Funding (8h) | Ann. Rate | Regime |
|---|---|---|---|---|---|---|
| BTC-USD-PERP | $X | X% | $X | X% | X% | Trending |
| ETH-USD-PERP | $X | X% | $X | X% | X% | Ranging |
| ... | ... | ... | ... | ... | ... | ... |

**Top Mover**: [MARKET] — [X%] in 24h
**Volume Leader**: [MARKET] — $X traded in 24h
**Highest Funding (8h)**: [MARKET] at X% — [longs/shorts] are paying [shorts/longs]
**Lowest Funding (8h)**: [MARKET] at X% — [longs/shorts] are paying [shorts/longs]

Funding Arbitrage Scan

Dedicated funding rate opportunity table:

## Funding Rate Scan — [date/time]

| Market | Current 8h Rate | Annualized | Direction to Collect | Persistent? | Notes |
|---|---|---|---|---|---|
| MARKET-A | 0.XX% | XX% | Short | 8/8 last periods | Strong |
| MARKET-B | -0.XX% | XX% | Long | 5/8 last periods | Moderate |

**Strongest opportunity**: [MARKET] — X% annualized by going [long/short].
Rate has been consistent for [N periods]. Watch for reversal if [condition].

Regime Summary

Current regime classification across all markets:

## Market Regime Summary

| Market | Regime | ADX | RSI | ATR vs Avg | Trend Direction |
|---|---|---|---|---|---|
| BTC-USD-PERP | Trending | 32 | 58 | 1.1x | Up |
| ETH-USD-PERP | Ranging | 15 | 48 | 0.8x | Neutral |
| SOL-USD-PERP | Volatile | 28 | 72 | 1.8x | Up |

**Overall market tone**: [risk-on / risk-off / mixed / uncertain]

Caveats

• Indicators are computed from available kline data — limited by API history depth

and resolution. Longer-period indicators (200 SMA) need sufficient data to be meaningful.

• Technical analysis is probabilistic, not predictive. Indicators describe current

conditions and historical tendencies, not future certainty.

• Funding rates can reverse quickly — a rate that looks attractive now can flip

direction within a single funding period.

• Orderbook depth is a snapshot in time. Large resting orders can be pulled

before they fill. Do not treat orderbook walls as guaranteed support/resistance.

• Cross-market screening shows current conditions. By the time you act, the data

may have shifted. Always re-check before entering.

• This is market analysis, not trading advice. Your job is to describe what the data shows,

not what the user should do with it. The distinction is subtle but important: state conditions, not implications.

• Neutral: "RSI is at 72, in overbought territory" — states a condition.
• Nudges toward a trade: "RSI at 72 suggests the rally may be exhausted" — implies action.
• Neutral: "Funding is 0.08%/8h; longs are paying shorts" — states a fact.
• Nudges toward a trade: "Elevated funding indicates crowded longs — a potential reversal setup" — implies action.
• Neutral: "Price has rallied 18% with volume 2.3x average" — states observation.
• Nudges toward a trade: "The move has conviction, worth watching for continuation" — implies action.

Present the data. Let the user decide what to do with it.

References

See [indicators.md](references/indicators.md) for detailed indicator formulas and parameter guidance.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: tradeparadex
• Source: tradeparadex/paradex-skills
• License: MIT

Install and usage instructions live in the source repository linked above.