edge-pipeline-orchestrator

Orchestrate multi-stage trading research pipelines from OHLCV detection to strategy export.

Updated Apr 17, 2026
One-click install
npx skills add https://github.com/pasie15/claude-trading-skills-marketplace --skill edge-pipeline-orchestrator-pasie15
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: edge-pipeline-orchestrator
Source: https://github.com/pasie15/claude-trading-skills-marketplace/tree/main/plugins/trading-strategy-tools/skills/edge-pipeline-orchestrator
Command: npx skills add https://github.com/pasie15/claude-trading-skills-marketplace --skill edge-pipeline-orchestrator-pasie15

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates and coordinates multi-stage trading research workflows so teams do not have to manually run detection, hint extraction, concept synthesis, drafting, review loops, and export steps across separate tools and scripts.

Core Features & Use Cases

  • Full pipeline orchestration: Run the pipeline from raw OHLCV or prebuilt tickets through hints, concepts, drafts, review-revision loops, and strategy export with a single command.
  • Resumable and safe runs: Resume from intermediate stages (drafts), perform dry-runs, and apply strict export rules to avoid exporting incomplete drafts.
  • LLM-augmented workflows: Ingest LLM-generated hints, promote ideas into concepts and drafts, and produce a pipeline_run_manifest.json for reproducibility.
  • Use Case: Run an end-to-end research job that auto-detects candidate anomalies from historical data, synthesizes trade concepts, designs draft strategies, iterates reviews, and exports PASSed strategies to strategy artifacts.

Quick Start

Run the edge pipeline from path/to/tickets and save outputs to reports/edge_pipeline/.

Frequently Asked Questions about edge-pipeline-orchestrator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate an end-to-end trading strategy research pipeline from raw OHLCV data?▼

You can automate the full pipeline from raw OHLCV data by orchestrating candidate detection, hint extraction, concept synthesis, draft design, review-revision loops, and strategy export with a single command. This coordinates all research stages automatically.

Can I resume a multi-stage edge research pipeline from an intermediate draft stage?▼

Yes, you can resume edge research pipeline runs from intermediate draft stages. The orchestration supports resuming from drafts, allowing you to continue review-revision loops and export without restarting the entire workflow.

How does LLM-augmented hint ingestion work in a trading strategy design workflow?▼

LLM-augmented hint ingestion allows you to feed external hints into the pipeline, promoting them into concepts and drafts. The workflow then designs strategies, iterates through review-revision loops, and writes a pipeline_run_manifest.json for reproducibility.

What is the best way to prevent exporting incomplete trading strategy drafts during automated research?▼

The best way to prevent exporting incomplete drafts is to use strict export rules within the pipeline orchestrator. This ensures only PASSed strategies from the review-revision loop are exported to strategy artifacts.

Does dry-run mode support pipeline orchestration for edge research without committing export artifacts?▼

Yes, dry-run mode supports pipeline orchestration without committing export artifacts. You can test the full detection, synthesis, and review-revision workflow safely before applying strict export rules to generate final strategy artifacts.