feature-store-for-live-and-backtest-parity SKILL.md skillA
feature-store-for-live-and-backtest-parity from himanshuj16/algo-trading-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add himanshuj16/algo-trading-skills/feature-store-for-live-and-backtest-parity@v1.1.0Pins v1.1.0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- himanshuj16/algo-trading-skills
- Path
skills/feature-store-for-live-and-backtest-parity/SKILL.md- Versions recorded
- 5
- Latest version
- v1.1.0, committed Sep 3, 2026
- First version recorded
- Jul 30, 2026
- Size
- 10,241 bytes
- Repository stars
- 50
- License
- Apache-2.0
Use when a model trains offline on bars and serves online on a stream, and two separate feature code paths would drift apart; one shared computation engine plus a numeric parity check between batch and streaming output.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
## When to Use Invoke this whenever a model is trained offline on historical bars and served online on a live market data stream. Writing one feature path for the backtest (Pandas vectorization) and a second for live trading (a stateful streaming loop) produces training-serving skew: the two agree on the day they are written and drift apart with every subsequent edit. Divergence in NaN handling, rolling-window boundaries, standard-deviation conventions, or warm-up behavior moves live predictions away from the ones the backtest was validated on, and the backtest keeps reporting the old, better numbers. The remedy is structural, not a reconciliation report: one shared calculation core called by both runtimes, plus an automated replay assertion that the two matrices agree. ## When NOT to Use - **Path-dependent features that are not window-derivable.** Wilder's RSI, EWMA volatility, and Kalman filters carry recursive state whose value depends on where the history was started. They cannot be reproduced from a bounded ring buffer. Serve them by checkpointing the smoothed state across the batch/online boundary — see …
Read the whole file at its exact version.
How to install
mdr add himanshuj16/algo-trading-skills/feature-store-for-live-and-backtest-parity@sha256:b86ff08d7b275836The line at the top of this page pins the label v1.1.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_kntweblqjkbp5cfx)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, between Jul 30, 2026 and Sep 3, 2026, about one every 9 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| v1.1.0 latest | 2026-09-03 | 8527a51 | 10,241 B | A | view · diff |
| v1.1.0 | 2026-09-03 | 64e8b01 | 10,238 B | A | view · diff |
| v1.1.0 | 2026-08-25 | 7e72405 | 10,186 B | A | view · diff |
| v1.0.0 | 2026-08-07 | 89d1436 | 3,712 B | A | view · diff |
| v1.0 | 2026-07-30 | dfc60c7 | 3,710 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 (10241 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
himanshuj16/algo-trading-skills · 50 stars · license Apache-2.0 · pushed 2026-09-14 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_kntweblqjkbp5cfx GET https://markdownregistry.com/api/v1/resolve?ref=himanshuj16/algo-trading-skills/feature-store-for-live-and-backtest-parity GET https://markdownregistry.com/api/v1/blob/b86ff08d7b2758366fe08e89692b821734d5334b0289c3c185f50457e7a0ab21
Your AI assistant 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.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.