ml4t-triple-barrier skillA
ml4t-triple-barrier is agent-read markdown (skill) from ml4t/skills: Label trades using profit-target, stop-loss, and time barriers with volatility-adaptive thresholds. Use when creating supervised labels for financial time series..
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# Triple-Barrier Labeling
Fixed return thresholds ignore volatility - a 2% move is noise in crypto but a signal in treasuries. Triple-barrier labels adapt to the asset's current regime.
## The Problem
Naive binary labels (`return > 0`) are noisy and ignore position management. A trade that gains 5% then gives back 8% is labeled "winning" if you only check the endpoint. Triple-barrier labeling mirrors real trading: you exit when you hit a profit target, a stop loss, or time runs out.
## The Pattern
### WRONG
```python
import numpy as np
# Fixed threshold ignores volatility regime
labels = np.where(fwd_returns > 0.02, 1, np.where(fwd_returns < -0.01, -1, 0))
```
### CORRECT
```python
import numpy as np
def triple_barrier_labels(
prices: np.ndarray,
upper_mult: float = 2.0,
lower_mult: float = 1.5,
atr_period: int = 14,
max_holding: int = 10,
) -> np.ndarray:
"""Label each bar: +1 profit hit, -1 stop hit, 0 time expiry."""
# Volatility-adaptive barriers via a TRAILING mean of absolute price changes.
# mode="same" would centre the window and let atr[i] see bars after i.
abs_changes = np.abs(np.diff(prices, prepend=prices[0]))
…Read the whole file at its exact version.
How to install
mdr add ml4t/skills/ml4t-triple-barrier@git:20260901.14ceba9mdr add ml4t/skills/ml4t-triple-barrier@sha256:d3ba460d283f7cb5Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_tdpi3evhobpp6tl7)
1 badge views in 30 days
Versions
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260901.14ceba9 latest | 2026-09-01 | 14ceba9 | 4,768 B | A | view · diff |
| git:20260901.c415df0 | 2026-09-01 | c415df0 | 4,542 B | A | view · diff |
| git:20260901.5ec4327 | 2026-09-01 | 5ec4327 | 4,552 B | A | view · diff |
| git:20260528.303089e | 2026-05-28 | 303089e | 4,484 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (4768 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
ml4t/skills · 19 stars · license Apache-2.0 · pushed 2026-09-24 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_tdpi3evhobpp6tl7 GET https://markdownregistry.com/api/v1/resolve?ref=ml4t/skills/ml4t-triple-barrier GET https://markdownregistry.com/api/v1/blob/d3ba460d283f7cb5982ca30dc2e34b5b88ada11c9b9c89c334a1b22d1f9ec970
Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.