continuous-learning-pattern-extraction

name	continuous-learning-pattern-extraction
description	Extract and learn patterns from trading data continuously. Use when identifying recurring patterns, building adaptive models, or continuous improvement workflows.
version	1.0.0
allowed-tools	Read, Write, Edit, Bash

Continuous Learning Pattern Extraction

Source: https://mcpmarket.com/tools/skills/continuous-learning-pattern-extraction

Use When

• You need to apply this capability as part of trading research workflows (data, features, backtests, ML, reporting).
• You want a reproducible output that can be committed to this repo (code, configs, docs).

Inputs To Ask For

• Objective: what success looks like (metric, constraints, time horizon).
• Data: symbols, timeframe, sampling, data sources, and leakage risks.
• Constraints: compute budget, latency, interpretability, and deployment requirements.

Outputs

• A concrete plan (steps + checks).
• A minimal implementation sketch (files to create/change) and verification steps.
• If applicable: a risk checklist (leakage, overfitting, evaluation pitfalls).

Workflow

1. Restate the task in measurable terms.
2. Enumerate required artifacts (datasets, features, configs, scripts, reports).
3. Propose a default approach and 1-2 alternatives.
4. Add validation gates (walk-forward, Monte Carlo, sanity checks).
5. Produce repo-ready deliverables (code + docs) and a run command.