news-trading

Name: news-trading
Author: SKE-Labs

15stars2forks

How to install news-trading?

ske-labs/agent-trading-skills/news-trading

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

Installs into the current project.

Summary

Plan and evaluate trades around scheduled economic or corporate releases. Use when the user needs primary-source event data, standardized surprise measurement, scenario triggers, and execution-risk controls.

Files · 1

View on GitHub

SKILL.md

1	# News Trading Strategy
2
3	Trade volatility from scheduled economic events using deviation-from-consensus scoring.
4
5	## Setup Conditions
6
7	### Event Inputs
8
9	Use a live official calendar, release timestamp and revision policy. Record every economically relevant component: for example, headline and core inflation; payrolls, unemployment, wages, and revisions; or a policy decision, statement, projections, and press conference. Do not attach a universal price move to an event name.
10
11	### Deviation-from-Consensus Scoring
12
13	```
14	Standardized surprise = (Actual - consensus) / historical standard deviation of surprises
15	```
16
17	Use a rolling, vintage-consistent surprise history and state the consensus provider and cutoff time. If history is insufficient, show the raw surprise and estimate range without inventing a score. Direction also depends on priced expectations, revisions, other components, and policy regime.
18
19	## Workflow
20
21	### 1. Check Economic Calendar
22
23	```
24	get_economics_calendar(from_date=<start>, to_date=<end>, impact="high")
25	```
26
27	Identify upcoming high-impact events. Note dates, times, affected markets.
28
29	### 2. Research Consensus
30
31	```
32	get_financial_news(topic="<event> forecast consensus <month> <year>", max_results=15)
33	```
34
35	Find the current consensus, estimate range, cutoff timestamp, and prior value. Confirm the actual and revisions from the official release, not a headline or social post.
36
37	### 3. Plan Scenarios
38
39	Pre-plan multiple component combinations and observable price triggers. A headline beat need not imply one direction when revisions, core measures, guidance, or policy expectations conflict.
40
41	### 4. Post-Release Entry
42
43	Enter only when the selected execution gate is met: spread below a tested limit, quotes stable enough to size, and a closed-bar or order-book trigger. Calibrate any waiting period in bars for the specific market. Use a volatility-based stop with position size reduced to keep risk fixed; include gap, halt, rejection, and slippage scenarios.
44
45	### 5. Event Clustering
46
47	When releases overlap, attribute the move cautiously and wait until all scheduled components are available. If they conflict or the event timestamp/data vintage is uncertain, return `no trade`.
48
49	### 6. Report to Orchestrator
50
51	Event details, consensus vs actual, deviation score, recommended action, conviction level.
52
53	## Evidence and Validation
54
55	- Treat the setup as a testable hypothesis, not a prediction. Define thresholds, entry, invalidation, and exit before evaluating outcomes.
56	- Calibrate on the same instrument, venue, session, and timeframe. Use closed candles and a held-out or walk-forward sample; record every variant tried.
57	- 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.
58	- Return observed inputs, missing data, cost assumptions, entry, invalidation, exit, and a valid, watch, or no-trade status.
59	- Research basis: [Federal Reserve research](https://www.federalreserve.gov/econres/feds/macroeconomic-news-announcements-systemic-risk-financial-market-volatility-and-jumps.htm) finds more volatility jumps on announcement days and sensitivity to surprise and uncertainty.
60
61	## Key Rules
62
63	- Never use unreleased, embargoed, or suspected material nonpublic information.
64	- Do not hold through an event without an explicit event-risk mandate and gap-loss scenario.
65	- Size from the stress loss and executable stop distance, not a fixed percentage reduction.
66	- Skip conflicting clustered releases or failed liquidity gates.
67	- Model spread widening, slippage, rejected orders, halts, revisions, and timestamp latency.
68
69	## Related Skills
70
71	- breakout-trading -- post-news consolidation breakouts follow breakout principles
72	- gap-trading -- news events often produce gap opens

People who liked this also viewed

6 picks

Type

news-trading

15stars2forks

View on GitHub

How to install news-trading?

ske-labs/agent-trading-skills/news-trading

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

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# News Trading Strategy
2
3	Trade volatility from scheduled economic events using deviation-from-consensus scoring.
4
5	## Setup Conditions
6
7	### Event Inputs
8
9	Use a live official calendar, release timestamp and revision policy. Record every economically relevant component: for example, headline and core inflation; payrolls, unemployment, wages, and revisions; or a policy decision, statement, projections, and press conference. Do not attach a universal price move to an event name.
10
11	### Deviation-from-Consensus Scoring
12
13	```
14	Standardized surprise = (Actual - consensus) / historical standard deviation of surprises
15	```
16
17	Use a rolling, vintage-consistent surprise history and state the consensus provider and cutoff time. If history is insufficient, show the raw surprise and estimate range without inventing a score. Direction also depends on priced expectations, revisions, other components, and policy regime.
18
19	## Workflow
20
21	### 1. Check Economic Calendar
22
23	```
24	get_economics_calendar(from_date=<start>, to_date=<end>, impact="high")
25	```
26
27	Identify upcoming high-impact events. Note dates, times, affected markets.
28
29	### 2. Research Consensus
30
31	```
32	get_financial_news(topic="<event> forecast consensus <month> <year>", max_results=15)
33	```
34
35	Find the current consensus, estimate range, cutoff timestamp, and prior value. Confirm the actual and revisions from the official release, not a headline or social post.
36
37	### 3. Plan Scenarios
38
39	Pre-plan multiple component combinations and observable price triggers. A headline beat need not imply one direction when revisions, core measures, guidance, or policy expectations conflict.
40
41	### 4. Post-Release Entry
42
43	Enter only when the selected execution gate is met: spread below a tested limit, quotes stable enough to size, and a closed-bar or order-book trigger. Calibrate any waiting period in bars for the specific market. Use a volatility-based stop with position size reduced to keep risk fixed; include gap, halt, rejection, and slippage scenarios.
44
45	### 5. Event Clustering
46
47	When releases overlap, attribute the move cautiously and wait until all scheduled components are available. If they conflict or the event timestamp/data vintage is uncertain, return `no trade`.
48
49	### 6. Report to Orchestrator
50
51	Event details, consensus vs actual, deviation score, recommended action, conviction level.
52
53	## Evidence and Validation
54
55	- Treat the setup as a testable hypothesis, not a prediction. Define thresholds, entry, invalidation, and exit before evaluating outcomes.
56	- Calibrate on the same instrument, venue, session, and timeframe. Use closed candles and a held-out or walk-forward sample; record every variant tried.
57	- 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.
58	- Return observed inputs, missing data, cost assumptions, entry, invalidation, exit, and a valid, watch, or no-trade status.
59	- Research basis: [Federal Reserve research](https://www.federalreserve.gov/econres/feds/macroeconomic-news-announcements-systemic-risk-financial-market-volatility-and-jumps.htm) finds more volatility jumps on announcement days and sensitivity to surprise and uncertainty.
60
61	## Key Rules
62
63	- Never use unreleased, embargoed, or suspected material nonpublic information.
64	- Do not hold through an event without an explicit event-risk mandate and gap-loss scenario.
65	- Size from the stress loss and executable stop distance, not a fixed percentage reduction.
66	- Skip conflicting clustered releases or failed liquidity gates.
67	- Model spread widening, slippage, rejected orders, halts, revisions, and timestamp latency.
68
69	## Related Skills
70
71	- breakout-trading -- post-news consolidation breakouts follow breakout principles
72	- gap-trading -- news events often produce gap opens

People who liked this also viewed

6 picks

Type