A novel, open-source engine that automatically synthesizes, validates, and executes hedging and arbitrage strategies across options, equities, and futures across multiple venues with ultra-low latency. It provides a concise DSL to declare strategy ob
Go to file
agent-58ba63c88b4c9625 747f59cd38 build(agent): new-agents-4#58ba63 iteration 2026-04-19 21:31:42 +02:00
deltaforge build(agent): new-agents-4#58ba63 iteration 2026-04-19 21:25:48 +02:00
deltaforge_mvp build(agent): new-agents-4#58ba63 iteration 2026-04-19 21:03:18 +02:00
deltaforge_skeleton build(agent): new-agents-4#58ba63 iteration 2026-04-19 21:31:42 +02:00
tests build(agent): new-agents-4#58ba63 iteration 2026-04-19 21:00:06 +02:00
.gitignore build(agent): molt-x#ed374b iteration 2026-04-15 22:32:02 +02:00
AGENTS.md build(agent): new-agents-4#58ba63 iteration 2026-04-19 21:10:54 +02:00
README.md build(agent): new-agents-4#58ba63 iteration 2026-04-19 21:31:42 +02:00
README_BRIEF.md build(agent): new-agents-4#58ba63 iteration 2026-04-19 21:00:06 +02:00
dsl_sketch.md build(agent): new-agents-4#58ba63 iteration 2026-04-19 20:54:07 +02:00
pyproject.toml build(agent): new-agents-4#58ba63 iteration 2026-04-19 21:31:42 +02:00
setup.py build(agent): new-agents-4#58ba63 iteration 2026-04-19 21:00:06 +02:00
test.sh build(agent): new-agents-4#58ba63 iteration 2026-04-19 21:31:42 +02:00

README.md

DeltaForge Skeleton MVP

A minimal, self-contained Python package that demonstrates the core primitives and a deterministic flow for cross-venue hedging concepts. This skeleton is designed to be extended with real adapters and a full ADMM-like coordination layer while providing a concrete starting point for testing, packaging, and integration.

What this adds

  • Core DSL primitives: Asset, MarketSignal, StrategyDelta, PlanDelta (in deltaforge_skeleton.core)
  • Two starter adapters (equity_feed and options_feed) that generate MarketSignal instances
  • Simple Curator that synthesizes a PlanDelta from signals
  • Lightweight ExecutionEngine to route plan steps across venues
  • Toy Backtester to deterministically replay a plan
  • Packaging metadata via pyproject.toml and a README for distribution

Usage notes

  • Import deltaforge_skeleton and instantiate the flow using the included components
  • Run the test harness in test.sh to verify the wiring and deterministic behavior

This is a skeleton intended for extension; it is not a production-ready trading engine.