Generator

Trading System Blueprinter

Pick your data source, LLM, broker, storage, risk engine, logger. Get a Mermaid architecture diagram + starter repo scaffold (ZIP). Browser-only. Free.

AI Fin Hub Research Published Apr 20, 2026 Methodology Corrections

Inputs: Configuration
Runtime: Instant
Privacy: Client-side · no upload
API key: Not required
Methodology: Open →

1 · Pick your stack

Data source

Where your bars / quotes / book events come from.

Broker

Who your orders route through.

LLM layer

Research + decision-support model. BYO keys.

Storage

How you persist bars, trades, decisions, dossiers.

Risk engine

What stops you from blowing up.

Logging + observability

How you see what the agent actually did.

Scheduler

How cadence is enforced.

2 · Architecture (Mermaid)

flowchart LR
  SCHED["launchd"] -->|tick| DATA["alpaca"]
  DATA -->|bars / book events| NORM[Normalize + persist]
  NORM --> STORE["duckdb"]
  STORE --> RESEARCH["claude-opus-4-7 · price-blind research"]
  RESEARCH -->|proposal + confidence| RISK["fractional-kelly"]
  RISK -->|sized order| EXEC["alpaca"]
  EXEC -->|fills + errors| LOG["heartbeat-json"]
  LOG --> STORE
  LOG -.-> ALERT[Telegram / email]

Paste into any Mermaid-compatible renderer (Notion, Obsidian, GitHub, VS Code preview).

3 · Starter repo scaffold

trading-system/
├── README.md                    # start here
├── .env.example                 # API keys (never commit real values)
├── scripts/
│   ├── fetch-bars.py            # data source → local duckdb
│   ├── research.py              # LLM call (price-blind context builder)
│   ├── decide.py                # risk sizing + decision logger
│   ├── execute.py               # broker call (idempotent)
│   └── heartbeat.py             # health probe + alert
├── data/
│   ├── heartbeat.json           # last-heartbeat timestamp (watchdog reads)
│   ├── circuit.json             # {"paused": false, "reason": ...}
│   └── tickers/                 # per-ticker markdown dossiers
├── memory/
│   └── decisions.jsonl          # append-only decision log
├── plists/
│   └── com.you.trader-*.plist   # launchd-based schedules
├── tests/
│   ├── test_research_no_price_leak.py
│   ├── test_sizing_caps.py
│   └── test_idempotent_orders.py
└── pyproject.toml

Minimal, opinionated layout. Everything non-runtime lives outsidescripts/. Add pyproject / requirements to taste.

Why this shape

· Schedule → data → research → risk → execute → log enforces the ordering that keeps your LLM out of the price decision loop.
· Heartbeat + circuit breaker are first-class so you can spot dead pipelines in under one cycle.
· Append-only decisions.jsonl makes post-mortems possible; any LLM-driven system is unauditable without it.
· Tests cover the three most dangerous failure modes: price leakage, sizing blow-up, duplicate orders.

See methodology for the principle set this generator encodes.

Complementary tools

Finance MCP Directory

Security-graded catalog of finance MCP servers — Alpaca, Polygon, Databento, IBKR, Tradier, Tiingo, NautilusTrader. Scope, auth, idempotency, transport, schema quality, all in one place.

Directories Open

Data-Vendor TCO Calculator

Compute annual cost of market data across Databento, Polygon, Alpaca, Tiingo, FMP, and Alpha Vantage for your exact universe, bar resolution, history depth, and API call volume.

Comparators Open

Price-Blind Research Auditor

Paste a research prompt or agent context bundle. The auditor flags price numbers, directional words, and outcome-leaking phrases that cause LLMs to retroactively rationalize positions. Builds a price-blind research boundary.

Playgrounds Open