Edge Pipeline Orchestrator

Orchestrates a complete edge research pipeline from market data to exportable trading strategies, handling candidate detection, edge synthesis, strategy drafting, and iterative review/revision cycles. You'd use this when you want to automate the full workflow instead of running each stage manually. It can start from raw OHLCV data or existing tickets, resume from any stage, and includes a review loop that passes, rejects, or revises drafts over up to two iterations. The LLM augmented mode is interesting: you analyze market data in Claude Code, write hints to YAML, and let the orchestrator promote those to full strategy concepts. Outputs a complete manifest with all artifacts organized by stage. Saves time if you're running this pipeline repeatedly, though the multi-stage setup assumes you've already built out the underlying edge research tooling.

skills.sh installs: 3.4K
GitHub stars: 3.7K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add tradermonty/claude-trading-skills --skill edge-pipeline-orchestrator --agent claude-code

Installs into .claude/skills of the current project.

Files

Edge Pipeline Orchestrator

Coordinate all edge research stages into a single automated pipeline run.

When to Use

Run the full edge pipeline from tickets (or OHLCV) to exported strategies
Resume a partially completed pipeline from the drafts stage
Review and revise existing strategy drafts with feedback loop
Dry-run the pipeline to preview results without exporting

Prerequisites

API keys: None required. The orchestrator runs the edge pipeline subskills through subprocess calls.
Sibling skills: The orchestrator resolves skills/edge-candidate-agent, skills/edge-hint-extractor, skills/edge-concept-synthesizer, skills/edge-strategy-designer and skills/edge-strategy-reviewer by repository path. Install or check out all of them together; a standalone install of this skill alone will not run.
Python: 3.9+ with PyYAML>=6.0 (see requirements.txt).
Inputs: A tickets directory, an OHLCV CSV for --from-ohlcv, or an existing drafts directory, depending on the mode shown under CLI Usage.