circuit-framework

What it is

Circuit Framework is a crypto-native multi-agent research and paper-trading framework where specialized agents analyze market structure, derivatives, sentiment, catalysts and market regime before a deterministic risk engine approves or rejects each trade. It is a fork of TradingAgents and keeps the internal package name tradingagents for compatibility; branding and crypto workflows are branded as Circuit Framework. The project explicitly states: "Research only — not financial advice. LLM output can be incorrect. Paper results do not represent live execution. No real trades are placed. This software never calls the Hyperliquid Exchange endpoint and never requests wallet credentials."

How it works

The architecture flows from a crypto analyze trigger to a Snapshot Builder, then through Market Structure, Derivatives, Sentiment, Catalyst, and Regime, culminating in Bull/Bear Researchers and a Research Manager that emits a Trade Proposal to a Deterministic Risk Gate. The risk gate sizes, clamps, and may reject before Paper Execution or a Paper Report Bundle. All analysts share a single immutable CryptoMarketSnapshot built at the start of the run. The workflow is defined via YAML profiles (balanced, momentum, mean_reversion, derivatives, narrative, macro_regime, quant_systematic).

Getting started

Installation and usage examples provided in the README include:

• Paper trading CLI sequences:

  pip install -e ".[dev]"
  
  tradingagents crypto analyze BTC
  tradingagents crypto analyze ETH --strategy momentum
  tradingagents crypto analyze SOL --interval 1h
  tradingagents crypto analyze HYPE --strategy derivatives --paper
  tradingagents crypto portfolio
  tradingagents crypto positions
  tradingagents crypto leaderboard
  

• Programmatic crypto run:

  from tradingagents.graph.trading_graph import TradingAgentsGraph
  from tradingagents.default_config import DEFAULT_CONFIG
  from tradingagents.graph.setup import CRYPTO_DEFAULT_ANALYSTS
  
  config = DEFAULT_CONFIG.copy()
  ta = TradingAgentsGraph(
      selected_analysts=list(CRYPTO_DEFAULT_ANALYSTS),
      config=config,
      asset_type="crypto",
      strategy_profile="balanced",
  )
  state, decision = ta.propagate(
      "BTC",
      "2026-07-14",
      asset_type="crypto",
      strategy_profile="balanced",
  )
  

• Environment variables and testing commands are documented, including offline crypto tests and fixture usage.

Getting started (install and config)

• Installation steps:

  git clone <this-repo>
  cd circuit-framework
  python3 -m pip install -e ".[dev]"
  cp .env.example .env   # add LLM API keys for live analysis
  

• Docker guidance exists: docker compose run --rm tradingagents.

Recent releases

• Releases section shows no entries (latest 0).

Traction

• Stars: 487
• Stars in last 24h: 1

Behind the repo

• This repo is a fork of TradingAgents by Tauric Research and maintains compatibility via the internal package name tradingagents.
• Upstream attribution cites Apache 2.0 license and a connection to TradingAgents.

Caveats

• License: Apache-2.0 (as upstream attribution notes)
• Created: 2026-07-16
• Last push: 2026-07-16
• Known limitations documented include: on-chain data not included until a verified provider; LLM analysts can be wrong; paper fills use mid ± slippage and configured fees; data sources are offline for tests.