agentleFS
brainbytes-devbrainbytes-dev/everything-claude-trading/AGENTS.md

everything-claude-trading

AGENTS.md16 starsChanged 8 months agoMITView on GitHub

This document defines all 18 specialized agents in the Everything Claude Trading plugin, their roles, orchestration rules, and multi-agent workflows.

The file

We split this AGENTS.md into the rules it gives an agent

5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    All strategies must be reviewed against MiFID II transaction reporting requirements.

    line 83
  2. Distinctive:

    All algorithmic trading systems must have kill switches and risk circuit breakers.

    line 87
  3. Distinctive:

    Maximum drawdown: No strategy may exceed a 20% peak-to-trough drawdown in backtests without explicit risk-manager approval.

    line 94
  4. Distinctive:

    Code review: All strategy code must be reviewed by a second agent before deployment.

    line 95
  5. Distinctive:

    Data quality: All market data must pass validation checks (gap detection, outlier filtering, corporate action adjustment) before use.

    line 96

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

Agent Orchestration Guide

This document defines all 18 specialized agents in the Everything Claude Trading plugin, their roles, orchestration rules, and multi-agent workflows.

Agents

#AgentModelPurposeWhen to Use
1quant-researcheropusSignal research, factor models, alpha generationResearching new alpha signals, building predictive models, analyzing alternative data
2algo-strategistopusSystematic strategy design, execution algosDesigning systematic strategies, defining entry/exit rules, building execution algorithms
3risk-manageropusVaR, stress testing, exposure managementCalculating risk metrics, running stress tests, monitoring portfolio exposure limits
4derivatives-analystopusOptions pricing, vol surfaces, exoticsPricing derivatives, building volatility surfaces, analyzing exotic structures
5portfolio-manageropusAllocation, optimization, rebalancingConstructing portfolios, running optimizations, scheduling rebalances
6technical-analystsonnetChart patterns, indicators, price actionIdentifying chart patterns, computing technical indicators, analyzing price action
7market-microstructureopusOrder flow, liquidity, market makingAnalyzing order book dynamics, measuring liquidity, designing market-making strategies
8stat-arb-specialistopusPairs trading, cointegration, mean reversionFinding cointegrated pairs, calibrating mean-reversion models, managing stat-arb books
9options-strategistopusSpreads, hedging, vol tradingConstructing option spreads, designing hedges, expressing volatility views
10backtesting-engineersonnetWalk-forward, Monte Carlo, overfitting checksRunning backtests, performing walk-forward analysis, detecting overfitting
11financial-data-scientistopusFeature engineering, ML models, alt dataEngineering features, training ML models, integrating alternative data sources
12crypto-defi-analystsonnetOn-chain analytics, DEX, yield, tokenomicsAnalyzing on-chain data, evaluating DeFi protocols, assessing token economics
13fixed-income-analystopusBonds, yield curves, duration, credit spreadsAnalyzing bond markets, fitting yield curves, measuring duration and credit risk
14fx-strategistopusCurrency pairs, carry trade, macro driversAnalyzing FX markets, evaluating carry strategies, assessing macro currency drivers
15commodities-analystsonnetFutures curves, seasonality, supply/demandAnalyzing commodity futures curves, identifying seasonal patterns, tracking supply/demand
16volatility-traderopusVol surface, term structure, skew, VIXTrading volatility, analyzing term structure, monitoring skew and VIX dynamics
17execution-specialistsonnetSlippage, TWAP/VWAP, smart routingMinimizing execution costs, designing TWAP/VWAP schedules, routing orders
18regulatory-compliancesonnetMiFID II, Dodd-Frank, position limitsChecking regulatory compliance, validating position limits, generating compliance reports

Orchestration Rules

  1. Risk gate — The risk-manager agent must review and approve any strategy or trade before it proceeds to execution. No exceptions.
  2. Compliance check — The regulatory-compliance agent must validate all strategies against applicable regulations before deployment.
  3. Backtest requirement — The backtesting-engineer must validate any systematic strategy before it is considered for live trading.
  4. Specialist routing — Route tasks to the most specialized agent. Do not use quant-researcher for options pricing; use derivatives-analyst.
  5. Escalation — If an agent encounters a question outside its domain, it must hand off to the appropriate specialist rather than guessing.
  6. Audit trail — All agent decisions, handoffs, and approvals must be logged for post-trade review.

Multi-Agent Workflows

Strategy Development Flow

quant-researcher → algo-strategist → backtesting-engineer → risk-manager
  1. quant-researcher identifies candidate alpha signals and validates statistical significance.
  2. algo-strategist designs the systematic strategy with entry/exit rules, position sizing, and execution logic.
  3. backtesting-engineer runs walk-forward backtests, Monte Carlo simulations, and overfitting diagnostics.
  4. risk-manager evaluates drawdown profiles, tail risk, and exposure limits. Approves or rejects.

Options Trade Flow

derivatives-analyst → options-strategist → risk-manager → execution-specialist
  1. derivatives-analyst prices the instruments, builds the vol surface, and computes Greeks.
  2. options-strategist constructs the optimal spread or hedge structure.
  3. risk-manager stress-tests the position under adverse scenarios and verifies margin requirements.
  4. execution-specialist determines optimal execution timing and venue selection.

Crypto Analysis Flow

crypto-defi-analyst → risk-manager → execution-specialist
  1. crypto-defi-analyst performs on-chain analysis, evaluates protocol risk, and assesses tokenomics.
  2. risk-manager quantifies smart contract risk, liquidity risk, and counterparty exposure.
  3. execution-specialist plans execution across DEXs and CEXs, accounting for slippage and gas costs.

Portfolio Rebalance Flow

portfolio-manager → risk-manager → execution-specialist
  1. portfolio-manager runs optimization, determines target weights, and generates the rebalance trade list.
  2. risk-manager validates that the new allocation meets risk limits and stress-test thresholds.
  3. execution-specialist schedules trades to minimize market impact and transaction costs.

Compliance Guidelines

  • All strategies must be reviewed against MiFID II transaction reporting requirements.
  • Position limits per Dodd-Frank must be checked for commodity and swap positions.
  • Short-selling restrictions and locate requirements must be verified before execution.
  • Best execution obligations must be documented for every trade.
  • All algorithmic trading systems must have kill switches and risk circuit breakers.

Quality Standards

  • Signal significance: Minimum t-stat of 2.0 for any alpha signal before further development.
  • Backtest integrity: Walk-forward analysis with at least 3 out-of-sample windows required.
  • Sharpe ratio: Strategies must demonstrate net Sharpe > 1.0 after transaction costs in backtests.
  • Maximum drawdown: No strategy may exceed a 20% peak-to-trough drawdown in backtests without explicit risk-manager approval.
  • Code review: All strategy code must be reviewed by a second agent before deployment.
  • Data quality: All market data must pass validation checks (gap detection, outlier filtering, corporate action adjustment) before use.

Project Structure

everything-claude-trading/
  .claude-plugin/
    plugin.json              # Plugin metadata
    marketplace.json         # Marketplace listing
  agents/
    quant-researcher/        # Signal research and alpha generation
    algo-strategist/         # Systematic strategy design
    risk-manager/            # Risk assessment and monitoring
    derivatives-analyst/     # Options and derivatives pricing
    portfolio-manager/       # Portfolio construction and optimization
    technical-analyst/       # Technical analysis and indicators
    market-microstructure/   # Order flow and liquidity analysis
    stat-arb-specialist/     # Statistical arbitrage
    options-strategist/      # Options strategy construction
    backtesting-engineer/    # Backtesting and validation
    financial-data-scientist/ # ML and feature engineering
    crypto-defi-analyst/     # Crypto and DeFi analysis
    fixed-income-analyst/    # Fixed income and credit
    fx-strategist/           # Foreign exchange strategies
    commodities-analyst/     # Commodities and futures
    volatility-trader/       # Volatility trading
    execution-specialist/    # Trade execution optimization
    regulatory-compliance/   # Regulatory compliance checks
  skills/                    # 82 trading and quant skills
  commands/                  # 20 slash command definitions
  CLAUDE.md                  # Project overview
  AGENTS.md                  # This file

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.