inalpha

Overview

Inalpha is a professional quant agent framework, governed by engineering discipline. It treats LLM agents not as black-box signal generators, but as code-writing collaborators bounded by hooks, permissions, plan-then-execute approval, and a one-shot signature on every order path.

Agents pick the factors that work now. Instead of a hard-coded indicator set, they rank factors by time-series Rank IC and surface the ones currently effective (factor.timing), then use that to back research and timing. Data itself is source-attributed by default — as_of-stamped and freshness-checked — so agents don't quietly reason on stale data.

Several capability lines sit on top of that harness:

• Factor lab + factor timing — agents formalize, compute, IC-test, multiple-testing-check, and register factors, and rank them by time-series Rank IC to time entries; every hypothesis is logged with author, timestamp, and the economic-story gate decision.
• Multi-perspective research — a deep dive convenes technical / fundamental / sentiment analysts, plus an optional panel of investing legends (Buffett / Lynch / Wood / Burry / Druckenmiller / Marks) for opposing views that feed a synthesis.
• Risk engine — declarative rules (notional caps, price deviation, drawdown veto) enforced at the HTTP boundary, not in prompts.
• Strategy evolution — LLMs mutate full Python source; three sandbox gates (AST audit, subprocess isolation, Strategy protocol contract) precede any candidate run; multi-objective fitness (Sharpe + Calmar − turnover − drawdown) so no metric can be gamed alone.
• Machine-approved orders (no direct LLM path) — order intents go trade.create_plan → approve → execute_plan with a single-use, TTL-bound approval_token; the LLM has no direct path to placing an order, and every step is logged into the audit trail.
• Inari Omikuji — a shrine fortune draw (playful easter egg) — undecided on direction? Cast a hexagram or draw a tarot card for a vantage outside the data; hard-walled from decisions, it can't touch risk, orders, or factors (see Core Capabilities §7).

The name combines Inari (the Japanese fox deity of prosperity) with alpha (the quant term for excess return) — a companion that reads your direction and keeps every step on the record.

Status: Inalpha is in alpha (Phase D-12 — factor-library closure: 79 factors with lineage & decay watch (alert-only, no auto-trim), a restricted-DSL factor-discovery L1, and a three-party research debate — on top of D-11 multi-market paper trading (cross-currency cash + a live runner that auto-runs promoted strategies on live bars), D-10 multi-market data, and D-9 LLM-authored strategies + risk engine). Read the code, weigh in on design — do not run this against real money (real-money trading is out of scope).

The Operator Console

The Operator Console (apps/dashboard) is the home base — a runtime dashboard that surfaces everything you'd otherwise have to ask the agent for, with a docked agent chat on the right. The shots below are the live console, running locally.

See Quick Start to bring the console up locally (pnpm dev → http://localhost:3001).

Design Principles

System Architecture

Three software layers over one data layer. A request flows down; results flow back up.

L1 · User entry. The Operator Console (apps/dashboard) is the home base, with a docked agent chat. The mastra dev playground is there for live trace; direct CLI tool calls still work.

L2 · Orchestration — packages/orchestration (Mastra · TypeScript). The one layer LLMs run in: a single orchestrator agent, wrapped in its harness — tools, hook/permission middleware, the in-memory plan store, conversation memory, and telemetry.

L3 · Kernel services — Python · FastAPI. Four independent, stateful processes, each owning one job:

L4 · Persistence & external. Postgres + TimescaleDB holds all time-series and business state. External venues span crypto, US / A-share / HK and other Asian & European single-name equities, global indices, and FRED macro — the orchestrator routes each venue automatically by