Skills longbridge-quant

📦

longbridge-quant

Name: longbridge-quant
Author: longbridge

v1.0.0 Content revision r2 Medium Risk ⚙️ External commands

Build and Evaluate Quantitative Strategies

Quantitative research requires consistent data, statistical methods, and validation. This skill structures Longbridge workflows for analysis, screening, optimization, execution, and hedging.

Supports: Claude Codex Code(CC)

📊 69 Adequate

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request

Review the Skillstore skill "longbridge-quant" from https://skillstore.io/skills/longbridge-longbridge-quant.md and its manifest at https://skillstore.io/api/skills/longbridge-longbridge-quant/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/longbridge-longbridge-quant.md Signed manifest GET /api/skills/longbridge-longbridge-quant/manifest Signed lockfile GET /api/skills/longbridge-longbridge-quant/lockfile

Test it

Using "longbridge-quant". Compare a possible pairs trade between two technology stocks.

Expected outcome:

A research report covering return correlation, cointegration evidence, hedge ratio, spread Z-score, half-life, signal thresholds, and execution caveats.

Using "longbridge-quant". Screen an index for value, quality, momentum, and low volatility.

Expected outcome:

A ranked table with factor Z-scores and composite scores.
A methodology note explaining weights and missing-data handling.
A summary of factor dispersion, coverage, and limitations.

Using "longbridge-quant". Optimize a moving-average strategy without overfitting.

Expected outcome:

A parameter comparison with in-sample metrics, out-of-sample metrics, walk-forward results, degradation analysis, and a clear overfitting assessment.

Security Audit

Medium Risk

v7 • 8/8/2026 Open versioned report

Most static findings mistake Markdown backticks or ordinary financial terminology for shell execution or reconnaissance. Remote indicator execution can disclose submitted scripts, and the hedging workflow accesses authenticated portfolio positions despite public-access metadata.

Files scanned

1,257

Lines analyzed

Review items

False positives ignored

Confirmed security concerns (2)

Medium

SKILL.md:10-13 references/hedging.md:28-29

Authenticated Portfolio Access Is Undeclared

The metadata says login is not required, while the hedging workflow reads current portfolio positions using trade-scope authentication. This mismatch can obscure access to sensitive financial data.

Both the public-access metadata and the trade-scope positions command are explicit. The contradiction is direct and affects permission transparency.

Low

references/quant-cli.md:39

Network reconnaissance

Executes the script server-side and returns the computed indicator/plot values as JSON. Scripts are

The workflow explicitly sends user-provided scripts to a remote server for execution. This is not reconnaissance, but it creates a real code-disclosure risk if scripts contain sensitive logic.

Capability review items (1)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

Medium

SKILL.md:107

Ruby/shell backtick execution

| `ModuleNotFoundError: sklearn` | Run `pip install scikit-learn` |

The troubleshooting instruction tells the agent to run pip install scikit-learn. Package installation executes third-party build or installation logic and should require user confirmation.

Risk Factors

⚙️ External commands (15)

references/factor-screen.md:60-64 references/quant-cli.md:74 references/quant-cli.md:85-95 SKILL.md:26 SKILL.md:52 SKILL.md:54-56 SKILL.md:56-58 SKILL.md:58-100 SKILL.md:100-106 SKILL.md:106-107 SKILL.md:107 SKILL.md:118-119 SKILL.md:119-120 SKILL.md:120-124 SKILL.md:124-133

Detected Patterns

Network reconnaissance

Audited by: codex View Audit History →

Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report

Copy report link

https://skillstore.io/skills/longbridge-longbridge-quant/audits/7?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/longbridge-longbridge-quant/security.svg)](https://skillstore.io/skills/longbridge-longbridge-quant?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/longbridge-longbridge-quant?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/longbridge-longbridge-quant/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/longbridge-longbridge-quant.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Academic citations (APA · BibTeX · CFF)

APA citation

longbridge. (2026). longbridge-quant security audit report (audit version 7) [Author version 1.0.0]. Skillstore. https://skillstore.io/skills/longbridge-longbridge-quant/audits/7

BibTeX citation

@techreport{longbridge-longbridge-longbridge-quant-2026,
  author = {longbridge},
  title = {longbridge-quant security audit report (audit version 7)},
  institution = {Skillstore},
  year = {2026},
  number = {7},
  url = {https://skillstore.io/skills/longbridge-longbridge-quant/audits/7},
  note = {Author version 1.0.0}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "longbridge-quant security audit report (audit version 7)"
version: "1.0.0"
type: report
authors:
  - name: "longbridge"
date-released: "2026-08-08"
url: "https://skillstore.io/skills/longbridge-longbridge-quant/audits/7"
identifiers:
  - type: other
    value: "skillstore:longbridge-longbridge-quant:audit:7"
    description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: High

Architecture

100

Maintainability

Content

Community

Spec Compliance

What You Can Build

Research a Trading Pair

Test cointegration, estimate a hedge ratio, and define entry and exit signals from historical prices.

Rank an Equity Universe

Screen index constituents and compare factor scores using valuation, momentum, quality, and volatility measures.

Validate a Strategy

Create parameter searches and walk-forward tests that separate in-sample and out-of-sample performance.

Try These Prompts

Explain a Quantitative Method

Explain [METHOD] for [SYMBOL OR UNIVERSE]. Define required data, calculations, assumptions, outputs, and major risks.

Screen a Stock List

Screen [SYMBOLS OR INDEX] using [FACTORS] and [THRESHOLDS]. Rank passing symbols, explain weights, identify missing data, and cite the data source.

Design a Backtest

Design a backtest for [STRATEGY] on [SYMBOLS] over [PERIOD]. Include signals, costs, position sizing, metrics, and out-of-sample validation.

Optimize and Stress Test a Portfolio Strategy

Optimize [STRATEGY] across [PARAMETER RANGES]. Use walk-forward validation, quantify degradation, model execution costs, test hedge scenarios, and flag overfitting.

Best Practices

Confirm symbols, markets, date ranges, and data frequency before analysis.
Include transaction costs, slippage, liquidity, and out-of-sample validation in every backtest.
Request consent before remote script submission or authenticated portfolio access.

Avoid

Do not present historical performance or model signals as guaranteed future returns.
Do not optimize parameters only on the full dataset without an untouched validation period.
Do not install packages, submit proprietary scripts, or read portfolio positions without approval.