v2.0.1 — Post-v2.0.0 hardening: honest backtest + audit-driven fixes

v2.0.1 — Post-v2.0.0 hardening: honest backtest + audit-driven fixes

⚠️ Simulation-only toy project. No live order routing. All outputs are not investment advice — see the disclaimer in the README.

A multi-agent code audit of the v2 stack drove a round of correctness fixes, honest validation, and de-duplication. The headline: the first out-of-sample backtest of the v2 signal/sizing stack — published with the candid result that it shows no demonstrated alpha vs buy-and-hold.

🔬 Honest validation (the highlight)

• Out-of-sample backtest (docs/backtest-v2-results.md) — bridges the v2 signal/sizing stack into the A-share backtest engine (no look-ahead) and reports an equity curve, Sharpe, drawdown, and a buy-and-hold baseline over 2019–2024 (14 sector-diverse names). Result: +25.5% vs +351.7% buy-and-hold, 0/14 — the stack is a drawdown-limiter that caps trend upside, not an alpha engine. Reproduce with scripts/backtest_v2_universe.py. The machinery is real; the inputs are unproven, and now measured.
• make demo — one-command offline backtest on a bundled sample (no Docker, API keys, or network).

🐛 Fixed

• Risk gates that were no-ops — preflight now reads the persisted circuit-breaker halt state (is_halted()) and enforces T+1 sellability; review_agent score cutoffs are named constants.
• Bayesian likelihood derivation — OutcomeTracker estimates genuine P(evidence | state) from conditional frequencies instead of mislabeling the hit-rate.
• T+N outcome horizons use trading days (not calendar days), so T+1 after a Friday no longer lands on a weekend.

🧹 Changed / consolidated

• Single source of truth for A-share constants (src/utils/ashare_constants.py).
• Unified data-source health on one shared DataSourceRouter (the /admin view was previously empty).
• Merged the two causal-chain engines into one (ImpactChainEngine → adapter over CausalChainConstructor); stock resolution and USD→gold kept lossless.
• LLM-debate de-risked with an opt-in graceful-degradation path (default off).
• Honest naming — Alpha158 → custom alpha factors; "causal chains" → rule-based templates; IC annotated as an autocorrelation proxy.
• Removed dead code and a no-op calibration dimension.

Quality

All external deps mocked in unit tests; full unit suite passes, ruff/format clean, CI green throughout. Every change shipped through the standard branch → PR → CI flow.

Full changelog: see CHANGELOG.md · grounded in ROADMAP.md.

v2.0.0 — AI-first autonomous agent architecture

v2.0.0 — AI-first autonomous agent architecture

⚠️ Simulation-only toy project. No live order routing. All outputs are not investment advice — see the disclaimer in the README.

Major architecture upgrade: the platform moves from a linear
data → analysis → prediction → strategy pipeline to an AI-first autonomous agent
centered on an OODA decision loop — fed by a market-intelligence pipeline, quant signals,
and a smart stock screener, over a Redis-Streams event bus.

Highlights

• Autonomous agent loop (src/agent_loop/) — OODA cycle: signal aggregation →
  Bayesian prescreen → bull/bear debate → risk gates → Kelly sizing → trade proposal,
  with a 7-team InvestmentDirector, sentiment-cycle gates, thesis tracking, T+1/T+3/T+5
  outcome tracking and confidence calibration.
• Market intelligence (src/intelligence/, src/intelligence_hub/) — 5-layer source
  hierarchy, 7-component scoring, causal impact chains, and a NetworkX knowledge graph.
• Quant & event bus (src/quant/, src/event_bus/) — HMM regime detection,
  YAML signal library, optional Qlib Alpha158, Redis-Streams event-driven micro-OODA.
• Risk & execution (src/risk/, src/trading/) — circuit breaker, VaR, Kelly sizing,
  kill switch, layered gates, A-share constraints (all simulation-only).
• 智能选股 / Smart stock recommendation (src/recommendation/) — multi-style screener +
  LLM review + T+1 overnight risk + win-rate tracking, with a user-facing UI and a live feed
  into the agent loop.
• Optional HTTP Basic Auth for the dashboard/API (WEB_USERNAME/WEB_PASSWORD; off by default).
• Docs rewritten to the v2 architecture (README, dev guide, CLAUDE.md, CHANGELOG).

Quality & security

• All external deps mocked in unit tests; 3801 unit tests pass, frontend type-check + build green.
• Imported as fresh, clean history — no upstream private state, session logs, or secrets.
• CodeQL triage: fixed real bugs + security findings (SSRF, path-injection, clear-text
  logging, ReDoS, stack-trace exposure), hardened CI workflows (least-privilege permissions,
  SHA-pinned actions); remaining low-severity findings triaged.

Merged via #39 (architecture + 智能选股 + Basic Auth) and #40 (CodeQL triage & hardening).

Full changelog: see CHANGELOG.md.

v0.1.0 — Initial public release

First open-source release of the A-share AI analysis platform — a personal learning / technical-exploration toy project (not investment advice; see the README disclaimer).

Highlights

• Data, analysis, prediction, strategy/backtest, and autonomous agent-loop layers
• FastAPI + React dashboard and a multi-model research workstation
• All credentials read from environment variables; clean git history (verified secret-free)

See CHANGELOG.md for the full feature list.