multi-timeframe-analysis

Name: multi-timeframe-analysis
Author: SKE-Labs

15stars2forks

How to install multi-timeframe-analysis?

ske-labs/agent-trading-skills/multi-timeframe-analysis

$npx -y skills add SKE-Labs/agent-trading-skills --skill multi-timeframe-analysis

Installs into the current project.

Summary

Map context, setup, and execution states across fixed timeframes. Use when measuring normalized EMA slope, objective structure, closed-bar triggers, and aligned or conflicting states without hand-built confidence scores.

Files · 1

View on GitHub

SKILL.md

1	# Multi-Timeframe Analysis (MTF)
2
3	Map context, setup, and execution states across fixed timeframes. Higher-timeframe information may help, but it does not automatically dominate; validate aligned and conflicting states separately.
4
5	## Timeframe Selection
6
7	\| Primary TF \| Higher TF \| Lower TF \| Use Case \|
8	\| ---------- \| --------- \| -------- \| ---------------- \|
9	\| 1D \| Weekly \| 4H \| Position trading \|
10	\| 4H \| 1D \| 1H \| Swing trading \|
11	\| 1H \| 4H \| 15m \| Intraday swing \|
12	\| 15m \| 1H \| 5m \| Day trading \|
13
14	Each timeframe should be 4-6x the one below it. Pick one combination and use consistently.
15
16	## Timeframe Roles
17
18	\| Role \| Purpose \| Focus \|
19	\| ----------- \| --------------- \| ---------------------------------------- \|
20	\| Higher \| Trend direction \| Major S/R, overall bias, regime \|
21	\| Primary \| Trade structure \| Patterns, setups, key levels \|
22	\| Lower \| Entry timing \| Precise entries, confirmation, tight stops \|
23
24	## State Output
25
26	Report each timeframe as bullish, bearish, ranging, or uncertain under its objective rule. Combine them as `aligned`, `mixed`, `counter-context`, or `incomplete`. Do not turn a hand-built score into confidence or position size; size comes from portfolio risk.
27
28	## Conflict Resolution
29
30	\| Conflict \| Resolution \|
31	\| --- \| --- \|
32	\| HTF bullish, Primary bearish \| Label mixed; evaluate the predeclared mixed-state policy. \|
33	\| HTF bearish, LTF bullish \| Label counter-context; trade only if that state is validated. \|
34	\| HTF ranging, Primary trending \| Label mixed; do not map to arbitrary size. \|
35	\| All timeframes conflicting \| No trade. Clarity is a prerequisite. \|
36
37	When evidence is incomplete or a state was not validated, return `no trade`.
38
39	## Workflow
40
41	1. Higher TF — establish bias via EMA stack + slope (no ADX — use the canonical slope computation). `get_indicators` returns one indicator per call — fetch each separately:
42	```
43	get_candles(symbol=<symbol>, exchange=<exchange>, interval=<htf_interval>, count=120)
44	get_indicators(indicator_code="ema_21", symbol=<symbol>, exchange=<exchange>, interval=<htf_interval>, count=120)
45	get_indicators(indicator_code="ema_50", symbol=<symbol>, exchange=<exchange>, interval=<htf_interval>, count=120)
46	get_indicators(indicator_code="atr_14", symbol=<symbol>, exchange=<exchange>, interval=<htf_interval>, count=120)
47	get_indicators(indicator_code="donchian_20", symbol=<symbol>, exchange=<exchange>, interval=<htf_interval>, count=120)
48	```
49	Then compute trend strength via slope %:
50	```
51	execute python3 -c "ema50=[<...>]; s=(ema50[-1]-ema50[-5])/ema50[-5]*100; print(f'htf_ema50_slope_pct={s:.3f}')"
52	```
53	Calibrate price/ATR-normalized slope bands in training data, then combine with the EMA stack. Absolute percentage thresholds are not portable across timeframes/assets.
54
55	Mark major HTF S/R: prior swing pivots, donchian_20 boundaries, Fibonacci extensions, weekly open / prior week high-low.
56
57	2. Primary TF — find setup aligned with HTF bias:
58	```
59	get_candles(symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
60	get_indicators(indicator_code="ema_9", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
61	get_indicators(indicator_code="ema_21", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
62	get_indicators(indicator_code="ema_50", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
63	get_indicators(indicator_code="rsi_21", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
64	get_indicators(indicator_code="macd_fast", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
65	get_indicators(indicator_code="atr_14", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
66	```
67	If HTF bullish → look for bullish setups (pullbacks to ema_21 / ema_50 / 50-61.8% Fib, demand zones, order blocks). Mark setup zones.
68
69	Cite the RSI / MACD slope as a 3–5 value progression (latest value alone is folklore):
70	- `RSI21 53.2 (49.8 → 51.4 → 52.9 → 53.2) rising` ← evidence of momentum continuation
71	- `MACD hist +8.4 (−1.2 → +3.6 → +8.4) accelerating bull` ← evidence of impulse
72
73	3. Lower TF — time the entry:
74	```
75	get_candles(symbol=<symbol>, exchange=<exchange>, interval=<ltf_interval>, count=20)
76	get_indicators(indicator_code="rsi_21", symbol=<symbol>, exchange=<exchange>, interval=<ltf_interval>, count=20)
77	get_indicators(indicator

People who liked this also viewed

6 picks

Type

multi-timeframe-analysis

15stars2forks

View on GitHub

How to install multi-timeframe-analysis?

ske-labs/agent-trading-skills/multi-timeframe-analysis

$npx -y skills add SKE-Labs/agent-trading-skills --skill multi-timeframe-analysis

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Multi-Timeframe Analysis (MTF)
2
3	Map context, setup, and execution states across fixed timeframes. Higher-timeframe information may help, but it does not automatically dominate; validate aligned and conflicting states separately.
4
5	## Timeframe Selection
6
7	\| Primary TF \| Higher TF \| Lower TF \| Use Case \|
8	\| ---------- \| --------- \| -------- \| ---------------- \|
9	\| 1D \| Weekly \| 4H \| Position trading \|
10	\| 4H \| 1D \| 1H \| Swing trading \|
11	\| 1H \| 4H \| 15m \| Intraday swing \|
12	\| 15m \| 1H \| 5m \| Day trading \|
13
14	Each timeframe should be 4-6x the one below it. Pick one combination and use consistently.
15
16	## Timeframe Roles
17
18	\| Role \| Purpose \| Focus \|
19	\| ----------- \| --------------- \| ---------------------------------------- \|
20	\| Higher \| Trend direction \| Major S/R, overall bias, regime \|
21	\| Primary \| Trade structure \| Patterns, setups, key levels \|
22	\| Lower \| Entry timing \| Precise entries, confirmation, tight stops \|
23
24	## State Output
25
26	Report each timeframe as bullish, bearish, ranging, or uncertain under its objective rule. Combine them as `aligned`, `mixed`, `counter-context`, or `incomplete`. Do not turn a hand-built score into confidence or position size; size comes from portfolio risk.
27
28	## Conflict Resolution
29
30	\| Conflict \| Resolution \|
31	\| --- \| --- \|
32	\| HTF bullish, Primary bearish \| Label mixed; evaluate the predeclared mixed-state policy. \|
33	\| HTF bearish, LTF bullish \| Label counter-context; trade only if that state is validated. \|
34	\| HTF ranging, Primary trending \| Label mixed; do not map to arbitrary size. \|
35	\| All timeframes conflicting \| No trade. Clarity is a prerequisite. \|
36
37	When evidence is incomplete or a state was not validated, return `no trade`.
38
39	## Workflow
40
41	1. Higher TF — establish bias via EMA stack + slope (no ADX — use the canonical slope computation). `get_indicators` returns one indicator per call — fetch each separately:
42	```
43	get_candles(symbol=<symbol>, exchange=<exchange>, interval=<htf_interval>, count=120)
44	get_indicators(indicator_code="ema_21", symbol=<symbol>, exchange=<exchange>, interval=<htf_interval>, count=120)
45	get_indicators(indicator_code="ema_50", symbol=<symbol>, exchange=<exchange>, interval=<htf_interval>, count=120)
46	get_indicators(indicator_code="atr_14", symbol=<symbol>, exchange=<exchange>, interval=<htf_interval>, count=120)
47	get_indicators(indicator_code="donchian_20", symbol=<symbol>, exchange=<exchange>, interval=<htf_interval>, count=120)
48	```
49	Then compute trend strength via slope %:
50	```
51	execute python3 -c "ema50=[<...>]; s=(ema50[-1]-ema50[-5])/ema50[-5]*100; print(f'htf_ema50_slope_pct={s:.3f}')"
52	```
53	Calibrate price/ATR-normalized slope bands in training data, then combine with the EMA stack. Absolute percentage thresholds are not portable across timeframes/assets.
54
55	Mark major HTF S/R: prior swing pivots, donchian_20 boundaries, Fibonacci extensions, weekly open / prior week high-low.
56
57	2. Primary TF — find setup aligned with HTF bias:
58	```
59	get_candles(symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
60	get_indicators(indicator_code="ema_9", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
61	get_indicators(indicator_code="ema_21", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
62	get_indicators(indicator_code="ema_50", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
63	get_indicators(indicator_code="rsi_21", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
64	get_indicators(indicator_code="macd_fast", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
65	get_indicators(indicator_code="atr_14", symbol=<symbol>, exchange=<exchange>, interval=<primary_interval>, count=60)
66	```
67	If HTF bullish → look for bullish setups (pullbacks to ema_21 / ema_50 / 50-61.8% Fib, demand zones, order blocks). Mark setup zones.
68
69	Cite the RSI / MACD slope as a 3–5 value progression (latest value alone is folklore):
70	- `RSI21 53.2 (49.8 → 51.4 → 52.9 → 53.2) rising` ← evidence of momentum continuation
71	- `MACD hist +8.4 (−1.2 → +3.6 → +8.4) accelerating bull` ← evidence of impulse
72
73	3. Lower TF — time the entry:
74	```
75	get_candles(symbol=<symbol>, exchange=<exchange>, interval=<ltf_interval>, count=20)
76	get_indicators(indicator_code="rsi_21", symbol=<symbol>, exchange=<exchange>, interval=<ltf_interval>, count=20)
77	get_indicators(indicator

People who liked this also viewed

6 picks

Type