technical-analysis

One good read of a chart is three things in order: where price is in its structure (trend and levels), what momentum and volatility are doing, and how the derivatives market is positioned around it. TraderSpy gives all three in at most two calls. Spend the calls well — the user's quota is finite — and quote the numbers, not adjectives.

Tools

Symbols are Binance Futures names: BTCUSDT, SOLUSDT, 1000PEPEUSDT (1000x-prefixed small caps), and tokenized equities/ETFs such as AAPLUSDT or SPYUSDT. Append USDT to a bare coin. Hyperliquid-only perps (e.g. HYPE) are tracked under the same naming.

Making the call count

• Multi-timeframe is one quota unit. For any "how does X look" question pass intervals: ["1h","4h","1d"] (or ["15m","1h","4h"] for an intraday user) and get a confluence verdict — never three separate calls.
• Several coins are one quota unit too. "How do BTC, ETH and SOL look" or a deeper read of a screen's top matches → symbols: ["BTCUSDT","ETHUSDT","SOLUSDT"] in ONE call, with the same intervals and indicators for all of them — never one call per coin. Up to three coins per call; for more, one call per three. For a quick side-by-side of many coins on one timeframe, screen_symbols (market-screener) is the better tool.
• Pick indicators for the question. Default rsi, macd, ema, bollinger covers momentum and trend. Add levels + pivots whenever the user cares about entries, targets or "where is support"; atr for stop distance and volatility; adx for trend strength; supertrend for a clean trend state; obv / mfi when volume confirmation matters; ichimoku / keltner only when asked. Eight to ten well-chosen indicators is a full read; all 19 is noise.
• periods lets you honour the user's settings ({"ema":[9,21,55,200],"rsi":7}); an unknown key is rejected, so use the documented names only.
• Add get_derivatives when the question involves "why is it moving", crowding, funding, open interest, leverage, liquidations, or when the user is about to lean on a directional call — candles cannot see positioning.
• get_candles is for when the user wants the data itself (to chart, export or compute something custom). Do not fetch candles to eyeball indicators the technicals tool already computes.

Reading the response

Shape. One timeframe → top-level indicators + summary. Several → timeframes[] (each with interval, price, candles, indicators, summary) plus confluence { aligned, bias, byInterval }; the top-level interval/price then name the first timeframe only. settings echoes what was computed; warnings[] lists anything that was thinned or missing. A symbols call returns data[] instead: one entry per coin, in the order you asked, each exactly the single-coin shape just described. A coin with no stored candles comes back as { symbol, error: "no_data", message } while the others still answer — say it was not available, do not drop it.

Every indicator carries value, previous, direction (rising / falling / flat) and a short series (oldest → newest), plus its own fields — zone on RSI/Stochastic/CCI/MFI/Williams, crossover on MACD/Stochastic, stack/pricePosition on EMA, ±DI and strength on ADX, trend/flipped/barsSinceFlip on SuperTrend, divergence on RSI/MACD/OBV. { value: null, reason: "insufficient_data" } means the tape is too short for that period — say so, do not substitute.

summary is the spine of the read. bias (bullish / bearish / neutral) is computed from a FIXED set (RSI 14, MACD, EMA 20/50/200, ADX 14, ATR 14, BB 20/2, volume 20) regardless of what you requested, so it is comparable between calls; score is its internal vote (|score| ≥ 30 sets a bias). trend (direction, strength, emaStack, priceVsEma200), momentum (rsiZone, macdState), volatility (atrPct, atrPercentile 0–100, squeeze, state), volume.ratioVsAverage, patterns[] (candlestick finders on the last two bars) and notes[] — plain-language sentences written to be quoted. Build the answer from the notes, then add the numbers.

levels (nearest three per side from swing highs/lows over 300 bars): price, touches, distancePct; range.positionPct (0 = at the low of the lookback, 100 = at the high); fibonacci (the dominant leg with retracedPct and the nearest level above/below); volumeProfile (poc, value area, pricePosition). pivots are classic floor pivots from the previous DAILY candle whatever the timeframe. Together they give you a levels table; distance in % is what a user can act on.

confluence.aligned is true only when every timeframe with data agrees on a non-neutral bias. "Aligned bearish on 1h/4h/1d" is a strong statement; "4h bullish, 1d neutral" is a pullback question. Say which.

Derivatives (get_derivatives): funding.label — extreme_longs_paying / longs_paying / neutral / shorts_paying / extreme_shorts_paying (longs pay at ≥ +0.03% per 8h, extreme at ≥ +0.10%; shorts pay at ≤ −0.01%, extreme at ≤ −0.05%; Binance's baseline is +0.01%; quote annualizedPct); openInterest.regime — new_longs / short_covering / new_shorts / long_liquidation / flat (needs ≥ 1% OI and ≥ 0.5% price change over 24h); positioning.label — long_heavy (long/short ratio ≥ 1.5, i.e. ≥ 60% long) / balanced / short_heavy (≤ 0.67, i.e. ≤ 40% long), read from the top-trader POSITION ratio first; takerBuySellRatio is the last hour's aggressor flow. Data is at most 60 s old. notes[] are quotable. Details in references/derivatives-guide.md.

Presenting

For a single-coin read ALWAYS use this order, and keep it to what the user asked:

1. Price line — symbol, price, time of the read, 24h change if you fetched it.
2. Per timeframe, one line each — bias, trend (EMA stack and ADX), momentum (RSI zone, MACD state), volatility (ATR %, squeeze), with the key number after each label.
3. Confluence — aligned or not, and what the disagreement means.
4. Levels table — nearest support and resistance with distance %, the pivot, the nearest Fibonacci level, the value-area edge if price is near it.
5. Derivatives line (if fetched) — funding label + annualized, OI regime, top-trader long %.
6. What would change the read — the one or two levels or readings that flip the picture (e.g. "a 4h close above 79,325 resistance with OI rising would turn the 4h bias bullish").
7. One plain sentence: crypto derivatives are high-risk; this is market information, not advice.

For several coins, give each coin steps 1–4 in short form, then one line comparing them (which are aligned, which disagree across timeframes, which sit closest to a level), and the risk sentence once at the end.

Use RSI(14) 41.3 — neutral, falling style labelling so the period and the direction are always visible. Percentages to one decimal; prices to the precision the tool returned.

Conduct

• Describe the chart; do not prescribe the trade. If the user asks "is this a good entry", answer with what supports the entry, what argues against it, where the invalidation sits, and hand the decision back. Never tell the user to buy, sell, size or leverage.
• Indicators describe the past bars they were computed on; do not present a bias as a forecast or imply any outcome is assured.
• Nothing in this connector trades — no order, close, transfer or withdrawal tool exists, by design. If asked to execute, say so plainly; the decision and the trade stay with the user.
• Quote only what the tools returned. If an indicator is insufficient_data or a symbol is not tracked, say so instead of estimating.
• When the answer is about a specific trade idea, end with one plain sentence that crypto derivatives are high-risk and this is market information, not financial advice — once per answer.

Per-indicator fields and conventional readings: references/indicator-guide.md.