Indicator Catalog

The engine ships 56 technical indicators (classic public indicators plus a small MobiusQuant proprietary set). Math is aligned with TradingView Pine Script. Each indicator has:

• A self-describing parameter signature (/api/indicators/registry)
• A drawing protocol for chart clients (/api/indicators/catalog)
• A knowledge entry — description, signal interpretation, caveats, analysis focus points (Chinese, see /zh/indicators/<id>)

Why this isn't just numbers — knowledge attached to every response

This is the key differentiator if you're building an AI Agent on top of indicators. Call /api/indicators?explain=true and the response carries the numerical series plus a knowledge block telling your LLM exactly how to read those numbers.

POST https://api.mobiusquant.ai/api/indicators?explain=true
  body: {"exchange":"binance","market":"perp","symbol":"BTCUSDT",
         "interval":"5m","limit":100,
         "calc":[{"name":"cvd"}]}

Response shape:

{
  "data": {
    "cvd_hist":  [-12, -8, -3, 5, 11, ...],
    "bear_div":  [false, false, true, ...],
    "bear_div_strength": [0, 0, 2, ...]
  },
  "knowledge": {
    "summary_focus": [
      "当前 cvd_hist 为正还是为负,反映买/卖力量主导",
      "最近是否触发了牛/熊背离,强度等级(Normal/Good/Strong)",
      "三个周期的背离信号是否存在共振"
    ],
    "signals": {
      "bear_divergence": "价格创新高但 cvd_hist 高点降低,多头量能不足,潜在空头反转",
      "strong_signal":   "强度=3(Strong),反转概率显著提升"
    },
    "caveats": "背离需要 fractal 确认延迟;单边趋势中信号较少;是反转预警而非即时信号"
  }
}

Why this matters for your LLM

Field	Source	What it does for your LLM
summary_focus	engine_indicator_knowledge.summary_focus	"Look at these things, in this order" — gives the LLM a structured walkthrough instead of free-form analysis
signals	.signals	Precise threshold definitions (e.g. RSI overbought = 70, not the LLM's guess) — prevents hallucinated rules
caveats	.caveats	Author-written pitfalls (e.g. fractal confirmation lag, fewer signals in single-direction trends) — prevents over-confident reads
guide_desc	.guide_desc	Use-case context — when this indicator is most informative

The result: your LLM stops guessing standard interpretive rules from training memory. It reads numbers + the rubric and produces grounded analysis. Hallucination drops noticeably; output structure becomes consistent across runs.

Recommended Skill pattern

Pass BOTH data and knowledge to your downstream LLM in a single prompt. Don't drop knowledge to save tokens — it's the part that makes the analysis trustworthy. See the Skill Runtime Lookup Guide for details.

---

Category breakdown

Category	Count	Examples
base	1	price
momentum	18	MACD, RSI family, TSI, ...
oscillator	4	RSI, Fisher Transform, RVI
trend	12	EMA, SuperTrend, PSAR, ...
trend_strength	4	ADX, Aroon, Vortex
volatility	10	ATR, Bollinger, Keltner, ...
volume	7	OBV, CMF, CVD, KVO, VWAP

Registry vs catalog

• /api/indicators/registry → full self-description (params, output cols, types) — consumed by programs
• /api/indicators/catalog → chartable subset + drawing metadata — consumed by chart clients
• /api/indicators → POST compute endpoint, returns klines + indicator values

---

Möbius Lab (proprietary)

ID	Category
mobius_trend	trend

These are MobiusQuant in-house indicators, not classic public ones.

---

All indicators

base (1)

ID
price

momentum (18)

ID
accelerator_oscillator
awesome_oscillator
cci
chande_mo
connors_rsi
coppock_curve
dpo
kst
macd
momentum
ppo
roc
smi_ergodic
stoch_rsi
trix
tsi
ultimate_oscillator
williams_r

oscillator (4)

ID
fisher_transform
rci_ribbon
rsi
rvi

trend (12)

ID
alligator
alma
ema
hma
lsma
mcginley_dynamic
mobius_trend
parabolic_sar
supertrend
tema
vwma
williams_stops

trend_strength (4)

ID
adx
aroon_oscillator
choppiness_index
vortex

volatility (10)

ID
atr
bollinger
chande_kroll_stop
donchian_channels
fibonacci_bollinger_bands
historical_volatility
keltner_channels
linear_regression_channel
mass_index
stdev

volume (7)

ID
chaikin_oscillator
cmf
cvd
kvo
obv
pvo
vwap

---

Usage examples

Single indicator: EMA(20) on BTC 5-minute

curl -X POST -H "Authorization: Bearer mq_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "exchange": "binance",
    "market":   "perp",
    "symbol":   "BTCUSDT",
    "interval": "5m",
    "limit":    100,
    "calc":     [{"name": "ema", "params": {"period": 20}}]
  }' \
  https://api.mobiusquant.ai/api/indicators

Multiple indicators: RSI + MACD + Bollinger

curl -X POST -H "Authorization: Bearer mq_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "exchange": "binance",
    "market":   "perp",
    "symbol":   "BTCUSDT",
    "interval": "1h",
    "limit":    200,
    "calc": [
      {"name": "rsi",       "params": {"period": 14}},
      {"name": "macd",      "params": {"fast": 12, "slow": 26, "signal": 9}},
      {"name": "bollinger", "params": {"period": 20, "stdev": 2}}
    ]
  }' \
  https://api.mobiusquant.ai/api/indicators

Same indicator with different params (use id to disambiguate)

{
  "calc": [
    {"id": "ema_fast", "name": "ema", "params": {"period": 12}},
    {"id": "ema_slow", "name": "ema", "params": {"period": 26}}
  ]
}

---

Full interpretation guide

The detailed signal interpretation, analysis focus points, and caveats for each indicator are documented in Chinese. Each English detail page links to its Chinese counterpart at the bottom.

For the rich Chinese version of this catalog: Chinese version.

Companion endpoints

Method	Path	Purpose
GET	/api/indicators/registry	All indicators — param signatures + output columns
GET	/api/indicators/catalog	Chartable subset + drawing protocol
POST	/api/indicators	Compute endpoint — returns klines + indicator values
POST	/api/chart	Same + drawing protocol for chart-rendering clients