momentum-trading

Name: momentum-trading
Author: SKE-Labs

15stars2forks

How to install momentum-trading?

ske-labs/agent-trading-skills/momentum-trading

$npx -y skills add SKE-Labs/agent-trading-skills --skill momentum-trading

Installs into the current project.

Summary

Define and test short-horizon directional momentum setups. Use when measuring normalized return, regime, participation, breakout/pullback entry, exhaustion exit, and transaction costs.

Files · 1

View on GitHub

SKILL.md

1	# Momentum Trading
2
3	Enter confirmed directional moves and ride them until momentum decays.
4
5	## Setup Conditions
6
7	### Regime Features
8
9	```
10	get_indicators(indicator_code="dmi", symbol=<symbol>, exchange=<exchange>, interval=<interval>)
11	```
12
13	Use ADX level and slope, normalized return, and relative volume as candidate features. Calibrate thresholds per instrument and timeframe, preferably against rolling percentiles; an ADX number is neither a probability nor a position-size instruction.
14
15	### Exit Features
16
17	- Volume declining while price advancing
18	- MACD histogram shrinking
19	- RSI divergence (price HH, RSI LH)
20	- Candles smaller with long wicks
21	- ADX turning down from its recent distribution
22
23	These features are correlated transformations of the same price/volume series. Define one exit rule and test it rather than counting two features as independent confirmation.
24
25	## Workflow
26
27	### 1. Check Regime
28
29	```
30	get_indicators(indicator_code="dmi", symbol=<symbol>, exchange=<exchange>, interval=<interval>)
31	```
32
33	Compare current ADX and normalized return with thresholds frozen from training data. If the regime gate fails, report `no trade` rather than substituting another strategy automatically.
34
35	### 2. Get Price Data and Confirm Momentum
36
37	```
38	get_candles_around_date(symbol=<symbol>, exchange=<exchange>, interval=<interval>, date=<date>)
39	get_indicators(indicator_code="rsi", symbol=<symbol>, exchange=<exchange>, interval=<interval>)
40	get_indicators(indicator_code="macd", symbol=<symbol>, exchange=<exchange>, interval=<interval>)
41	get_indicators(indicator_code="mfi", symbol=<symbol>, exchange=<exchange>, interval=<interval>)
42	```
43
44	Record multi-bar RSI and MACD progressions, time-of-day relative volume, and objectively detected swing structure. Avoid requiring every correlated indicator unless that conjunction improves held-out net expectancy.
45
46	### 3. Entry
47
48	Breakout: enter on a closed-bar break beyond a predeclared buffer. Pullback: define retracement depth as a tested interval rather than a fixed Fibonacci band. Invalidate beyond the structural level plus modeled slippage and size from the stop distance.
49
50	### 4. Report to Orchestrator
51
52	ADX value and trend direction, RSI/MACD/volume readings, entry type, exhaustion signals, stop and target levels.
53
54	## Evidence and Validation
55
56	- Treat the setup as a testable hypothesis, not a prediction. Define thresholds, entry, invalidation, and exit before evaluating outcomes.
57	- Calibrate on the same instrument, venue, session, and timeframe. Use closed candles and a held-out or walk-forward sample; record every variant tried.
58	- Include spread, fees, slippage, borrow or funding, partial fills, and latency. Reject the setup when net expectancy is not positive or depends on one narrow parameter.
59	- Return observed inputs, missing data, cost assumptions, entry, invalidation, exit, and a valid, watch, or no-trade status.
60	- Research basis: [Moskowitz & Grinblatt](https://onlinelibrary.wiley.com/doi/pdf/10.1111/0022-1082.00146) documented industry momentum, but implementation still requires regime, liquidity, cost, and crash-risk controls.
61
62	## Key Rules
63
64	- Do not map ADX directly to trade size or use a universal cutoff.
65	- Normalize extension and participation to the instrument, session, and timeframe.
66	- Predeclare whether entry is breakout or pullback; do not select retrospectively.
67	- Define the exit from observable price or feature conditions and test it after costs.
68	- Cap risk from the portfolio mandate, not from a momentum label.
69
70	## Related Skills
71
72	- breakout-trading -- breakout momentum is a primary entry type
73	- pullback-trading -- retracement entries use pullback principles for timing

People who liked this also viewed

6 picks

Type

momentum-trading

15stars2forks

View on GitHub

How to install momentum-trading?

ske-labs/agent-trading-skills/momentum-trading

$npx -y skills add SKE-Labs/agent-trading-skills --skill momentum-trading

Installs into the current project.

Summary

Define and test short-horizon directional momentum setups. Use when measuring normalized return, regime, participation, breakout/pullback entry, exhaustion exit, and transaction costs.

Files · 1

View on GitHub

SKILL.md

1	# Momentum Trading
2
3	Enter confirmed directional moves and ride them until momentum decays.
4
5	## Setup Conditions
6
7	### Regime Features
8
9	```
10	get_indicators(indicator_code="dmi", symbol=<symbol>, exchange=<exchange>, interval=<interval>)
11	```
12
13	Use ADX level and slope, normalized return, and relative volume as candidate features. Calibrate thresholds per instrument and timeframe, preferably against rolling percentiles; an ADX number is neither a probability nor a position-size instruction.
14
15	### Exit Features
16
17	- Volume declining while price advancing
18	- MACD histogram shrinking
19	- RSI divergence (price HH, RSI LH)
20	- Candles smaller with long wicks
21	- ADX turning down from its recent distribution
22
23	These features are correlated transformations of the same price/volume series. Define one exit rule and test it rather than counting two features as independent confirmation.
24
25	## Workflow
26
27	### 1. Check Regime
28
29	```
30	get_indicators(indicator_code="dmi", symbol=<symbol>, exchange=<exchange>, interval=<interval>)
31	```
32
33	Compare current ADX and normalized return with thresholds frozen from training data. If the regime gate fails, report `no trade` rather than substituting another strategy automatically.
34
35	### 2. Get Price Data and Confirm Momentum
36
37	```
38	get_candles_around_date(symbol=<symbol>, exchange=<exchange>, interval=<interval>, date=<date>)
39	get_indicators(indicator_code="rsi", symbol=<symbol>, exchange=<exchange>, interval=<interval>)
40	get_indicators(indicator_code="macd", symbol=<symbol>, exchange=<exchange>, interval=<interval>)
41	get_indicators(indicator_code="mfi", symbol=<symbol>, exchange=<exchange>, interval=<interval>)
42	```
43
44	Record multi-bar RSI and MACD progressions, time-of-day relative volume, and objectively detected swing structure. Avoid requiring every correlated indicator unless that conjunction improves held-out net expectancy.
45
46	### 3. Entry
47
48	Breakout: enter on a closed-bar break beyond a predeclared buffer. Pullback: define retracement depth as a tested interval rather than a fixed Fibonacci band. Invalidate beyond the structural level plus modeled slippage and size from the stop distance.
49
50	### 4. Report to Orchestrator
51
52	ADX value and trend direction, RSI/MACD/volume readings, entry type, exhaustion signals, stop and target levels.
53
54	## Evidence and Validation
55
56	- Treat the setup as a testable hypothesis, not a prediction. Define thresholds, entry, invalidation, and exit before evaluating outcomes.
57	- Calibrate on the same instrument, venue, session, and timeframe. Use closed candles and a held-out or walk-forward sample; record every variant tried.
58	- Include spread, fees, slippage, borrow or funding, partial fills, and latency. Reject the setup when net expectancy is not positive or depends on one narrow parameter.
59	- Return observed inputs, missing data, cost assumptions, entry, invalidation, exit, and a valid, watch, or no-trade status.
60	- Research basis: [Moskowitz & Grinblatt](https://onlinelibrary.wiley.com/doi/pdf/10.1111/0022-1082.00146) documented industry momentum, but implementation still requires regime, liquidity, cost, and crash-risk controls.
61
62	## Key Rules
63
64	- Do not map ADX directly to trade size or use a universal cutoff.
65	- Normalize extension and participation to the instrument, session, and timeframe.
66	- Predeclare whether entry is breakout or pullback; do not select retrospectively.
67	- Define the exit from observable price or feature conditions and test it after costs.
68	- Cap risk from the portfolio mandate, not from a momentum label.
69
70	## Related Skills
71
72	- breakout-trading -- breakout momentum is a primary entry type
73	- pullback-trading -- retracement entries use pullback principles for timing

People who liked this also viewed

6 picks

Type