A self-hosted investment decision engine built for modern AI agents. Multi-agent information isolation and cross-challenge protocol, providing an auditable decision trail (Audit Trail). Full Architecture Wiki · 中文版 OpenInvest is a self-hosted investment decision engine built for modern AI agents. It provides a verifiable investment committee, evidence-based reasoning, long-horizon backtest
npx mdskills install longsizhuo/openinvest@longsizhuo? Sign in with GitHub to claim this listing.Multi-agent investment committee with auditable decisions, backtesting, and transparent research methodology
1<div align="center">23<img width="120" height="120" alt="owl-02-lineart-gold" src="https://github.com/user-attachments/assets/e4c1efa0-026a-4777-ae62-48b9b0be435c" />45# openInvest67**A self-hosted investment decision engine built for modern AI agents. Multi-agent information isolation and cross-challenge protocol, providing an auditable decision trail (Audit Trail).**89[](https://www.python.org/)10[](docs/wiki/20-agent-usage-tutorial.md)11[](LICENSE)12[](https://github.com/longsizhuo/openInvest)13[](https://glama.ai/mcp/servers/longsizhuo/openInvest)1415[📚 Full Architecture Wiki](docs/wiki/README.md) · [🇨🇳 中文版](README_zh.md)1617</div>1819---2021## What is OpenInvest?2223OpenInvest is a self-hosted investment decision engine built for modern AI agents.2425It provides a verifiable investment committee, evidence-based reasoning, long-horizon backtesting, and auditable decision records. Instead of replacing Claude Code, Codex, Hermes, or OpenClaw, OpenInvest is designed to power them.2627---2829## Live Performance & PnL3031<div align="center">32 <img src="https://raw.githubusercontent.com/longsizhuo/openInvest/pnl-data/docs/pnl_chart.svg" alt="PnL chart" width="100%"/>33 <sub>Data feed is automatically updated every 2 hours using `jobs/pnl_snapshot` and pushed to the <a href="https://github.com/longsizhuo/openInvest/tree/pnl-data">pnl-data branch</a></sub>34 <br/>35 <sub>Upper half: 30-day net asset value trend · Lower half: Net asset value comparison against 8 benchmark assets (transparent disclosure, <b>not an alpha claim</b>—the committee's proven value is discipline and transparency, not excess return, see <a href="docs/wiki/adr/023-honest-positioning-not-alpha.md">ADR-023</a>)</sub>36 <br/>37 <sub>📌 <b>Note</b>: The current chart shows the author's live production portfolio. After self-hosting, the system will automatically render your own equity curve based on the holdings defined in your `memory/` directory.</sub>38</div>3940<!-- OUTPERFORM_FEED_START -->41<!-- OUTPERFORM_FEED_END -->4243* **Benchmark Portfolio**: The system introduces 8 standard control benchmarks across 4 quadrants (AI advisors / Mutual funds / Wealth management / Broad market index). For details on the comparison methodology and data cleaning logic, see [docs/wiki/README.md](docs/wiki/README.md).4445---4647## Research & Falsification4849**System Self-Disclosure**: This system is an **auditing tool to eliminate human investment cognitive biases and enforce reasoning transparency**, not a return-amplifying black box. Latest automated audit (`docs/verdict_accuracy.md`): Directional verdicts (excluding HOLD) have a true hit rate of **42.2%** (n=56, **below random**); `HOLD` accounts for **56%** of all decisions. The system's value lies in transparency and discipline (mostly staying inactive, low turnover), **not directional prediction**. Detailed log stream can be found in [docs/verdict_accuracy.md](docs/verdict_accuracy.md).5051This project systematically attempts to falsify its own edge and publishes negative results as-is. The deterministic features the committee reads, and the timing signals around them, were tested against pre-registered statistical gates — none survived as tradable alpha.5253| Test | Result | Verdict |54|---|---|---|55| Q1 cross-sectional stock picking | 6 features, mean-IC 0.025–0.067, Holm-corrected **p=0.397** | No significant stock-picking signal |56| M1 multivariate GBM (out-of-sample) | mean OOS IC **+0.003**, p=0.925 | Feature combination doesn't help either — no signal |57| Q2 gold MA200 trend | **p_holm=0.016**, significant — but `trend_dca` shows it is **beta, not tradable alpha**: timing terminal value **3.07 vs 15.10** buy-and-hold, Sharpe **+0.36 vs +0.68**, max drawdown deeper (**−57% vs −44%**) | Statistically significant, economically untradable |58| Per-asset multi-signal families | 3 assets × 4 signal families × parameter grid = 24 variants per asset; after costs + DSR deflation, **none passes DSR > 0.95** | No tradable signal in any family |59| Positive control | A cheating perfect-foresight timing signal scores **DSR = 1.00** | The harness can detect a real signal |6061Methodology: Newey-West HAC t-statistics, Deflated Sharpe Ratio (Bailey & López de Prado 2014, re-derived equation by equation), Holm correction, zero lookahead, and LLM training-cutoff probes.6263Details: [experiments/signal-eval/README.md](experiments/signal-eval/README.md) · [docs/verdict_accuracy.md](docs/verdict_accuracy.md) · [ADR-022](docs/wiki/adr/022-backtest-memory-contamination-and-holdout-discipline.md) · [ADR-023](docs/wiki/adr/023-honest-positioning-not-alpha.md)6465---6667## Product Philosophy6869Most AI investment assistants try to become better chatbots. OpenInvest instead builds a transparent, verifiable, and auditable decision engine that plugs into personal agents such as Claude Code, Codex, Hermes, and OpenClaw — every improvement in those agents automatically makes OpenInvest more capable.7071The division of labor is deliberate: your agent handles long-term memory, natural conversation, and user understanding; OpenInvest handles the verifiable investment committee, evidence-based reasoning, long-horizon backtesting, and auditable decision records.7273```74 User75 │76 ┌───────────┴───────────┐77 ▼ ▼78 Your Agent OpenInvest79(User Understanding) (Market Understanding)80 │ │81 └───────────┬───────────┘82 ▼83 Better Investment Decisions84```8586> **Your agent knows you. OpenInvest knows investing.**8788### Avoiding User Ownership89OpenInvest intentionally avoids "owning" the user. Most AI products try to own everything—memory, persona, chat history, and workspaces. OpenInvest takes a back seat. It exposes clean APIs, CLI commands, and agent skills (Claude Code / Codex / Hermes / OpenClaw), letting your primary agent manage the conversation and context while OpenInvest powers the underlying investment intelligence.9091---9293## Features9495* **Multi-Agent Investment Committee**: Isolated analysis and round-rebuttal debate.96* **Coordinator-Worker Architecture**: Prevents context contamination and role hallucination.97* **Information Isolation**: Rigidly blocks quant and risk analysts from out-of-boundary contexts.98* **Auditable Decision Trail**: Clean logs showing exactly "why" each decision was made.99* **Markdown-as-a-Database**: Frontmatter (YAML) + Markdown (Body) as the single source of truth.100* **Long-Horizon Backtesting**: Built-in test harness with lookahead bias protections.101* **Dreaming-Based Memory Consolidation**: Nightly memory distillation to prevent context drift.102* **Self-Hosted / Zero-Cost**: Powered directly by your local agent's reasoning resources.103* **Agent Skill**: Lightweight plugin for Claude Code / Codex / Hermes / OpenClaw, with interactive bootstrap wizard.104* **Automated Deployment**: GitHub Actions workflow to run the committee and email reports daily.105106---107108## Quick Start109110### 1. Integrate with your agent (Recommended)111Add the lightweight skill from your agent's plugin registry. The host agent will automatically pull the core code and align dependencies on first run:112```bash113# Claude Code114/plugin marketplace add longsizhuo/openInvest115/plugin install invest@openinvest116117# Codex118codex plugin marketplace add longsizhuo/openInvest119120# Hermes Agent121hermes plugins install longsizhuo/openInvest --enable122123# OpenClaw124openclaw plugins install clawhub:openinvest125```126Any other MCP client: register the MCP server from step 2 below (full walkthrough in the [agent tutorial](docs/wiki/20-agent-usage-tutorial.md)).127128### 2. Standalone — MCP server or CLI (no clone needed)129The backend ships on [PyPI](https://pypi.org/project/openinvest/); `~/openInvest` holds only your data:130```bash131# MCP (18 tools, any MCP client; add --http for a remote streamable-HTTP server — BETA)132claude mcp add openinvest -e INVEST_HOME=~/openInvest -- uvx openinvest-mcp133134# or plain CLI135INVEST_HOME=~/openInvest uvx openinvest status136```137Send `set up invest` (or `帮我初始化 invest`) to any skill-enabled AI terminal. The system will trigger an interactive bootstrap wizard to guide you through:1381. Detecting the `memory/` state storage path and `.env` configuration.1392. 5-dimensional profiling (Legal name, Risk capacity, Debt structure, Initial holdings, and optional keys).1403. Running static data migration to immediately generate your first asset exposure memo.141142> 💡 **Zero-Cost Execution**: In skill interactive mode, the committee's underlying reasoning relies entirely on the host agent's (e.g. Claude Code) reasoning pipeline. **No third-party API Key is consumed**. You only need to configure an API key when setting up automated crons or calling independent Web APIs.143144For self-hosting details, see [docs/QUICK_START.md](docs/QUICK_START.md). (The bundled Web GUI was retired on 2026-07-05 — all capabilities are exposed via CLI/MCP; a standalone frontend may return later.)145146### 3. Serverless Self-Hosting (GitHub Actions)147Run the committee automatically via GitHub Actions and receive daily digest emails.148> ⚠️ **Fork must be set to Private**: State files (holdings, verdicts) will be committed back to your fork. Public forks will leak your private financial information.1491501. **Fork this repository** and change its visibility to **Private** (Settings -> Visibility).1512. Run `set up invest` locally to generate the initial `memory/` folder, then commit and push it to your private fork:152 ```bash153 git add -f memory/ && git commit -m "chore: init memory state" && git push154 ```1553. In your fork's **Settings -> Secrets and variables -> Actions**, add the following Secrets:156 * `LLM_API_KEY` (or `DEEPSEEK_API_KEY`): API key to run the committee.157 * `EMAIL_SENDER` / `EMAIL_PASSWORD`: Gmail address + [App Password](https://support.google.com/accounts/answer/185833).158 * `DIGEST_EMAIL_TO`: Recipient email address.1594. **Enable Workflows** under the **Actions** tab. The workflow runs automatically at 10:00 AM (Beijing time) daily; you can also manually trigger `daily-report` via **Run workflow**.160161---162163## Architecture & Multi-Agent Orchestration164165openInvest does not run a mock debate in a single LLM session. The system enforces an **Information Isolation Contract** at the `core/committee/` layer, orchestrating 4 independent LLM processes in a directed acyclic graph (DAG):166167```168 [ Macro Data Injection ]169 │170 ▼ 1. Macro Alignment Context171 ┌──────────────────────────┐172 │ Macro Strategist │ (VIX / Interest rate spread / Currency momentum)173 └─────────────┬────────────┘174 │175 ▼ 2. Async Multi-Dimensional Scrutiny (Async DAG)176 ┌─────────────┴────────────┐177 ▼ ▼178 ┌──────────────────┐ ┌──────────────────┐179 │ Quant Analyst │ │ Risk Officer │180 │ (RSI / Momentum) │ │ (Concentration) │181 │ │ │ │182 │ 🛑 No Holdings │ │ 🛑 No Indicators │183 └─────────┬────────┘ └─────────┬────────┘184 │ │185 └─────────────┬─────────────┘186 │187 ▼ 3. Round 2 Rebuttal & Cross-Challenge188 │ Mutual feedback loop for signal correction189 ▼190 ┌──────────────────────────────────────────────┐191 │ Chief Investment Officer (CIO) │192 └───────────────────────┬──────────────────────┘193 │194 ▼ 4. Deterministic State Persistence195 [ BUY / ACCUMULATE / HOLD / TRIM / SELL ]196```1971981. **Macro Strategist**: Assesses the global macro landscape (VIX, yield curve spread, core currency matrix) to establish the portfolio's risk threshold.1992. **Quant Analyst**: A pure mathematical momentum and technical indicator filter. **Strictly blocked from knowing portfolio holdings** to eliminate human attachment and loss-aversion biases.2003. **Risk Officer**: Focuses entirely on tail risks (drawdown buffers, concentration limits, solvency multipliers). **Strictly blocked from technical indicators** to make objective asset exposure rulings.2014. **Round 2 Rebuttal**: Quant and Risk analysts are fed each other's Round 1 reports in Round 2, challenging boundaries until signals converge or safety valves trigger.2025. **CIO (Chief Investment Officer)**: Synthesizes the audited reports and outputs a structured `Verdict` (BUY / ACCUMULATE / HOLD / TRIM / SELL) with a confidence level. **No auto-order execution occurs**; final action remains strictly up to the human auditor.203204Key trade-offs behind this design are recorded as ADRs in [docs/wiki/adr/](docs/wiki/adr/) (24 to date), including rulings that overturned our own earlier designs — [ADR-007](docs/wiki/adr/007-few-shot-retirement.md) retired the few-shot CIO route, and [ADR-009](docs/wiki/adr/009-no-ta-style-analyst-agents.md) rejected TA-style analyst agents after a pre-registered experiment.205206---207208## Core Design209210* **Coordinator-Worker Pattern**: Workers operate in isolated namespaces. Boundary constraints are hardcoded at the framework layer in Python to prevent attention contamination in large multi-role prompts.211* **Markdown-as-a-Database**: The system uses Frontmatter (YAML) + Markdown (Body) as the single source of truth. Leveraging `fcntl.flock` process file locks and temporary atomic file replacement, it provides a tamper-proof investment audit trail natively tracked by Git.212* **Three-Phase Dreaming Consolidation**: Distills daily decisions against actual market outcomes nightly (Light Sleep $\rightarrow$ REM $\rightarrow$ Deep Sleep) to consolidate long-term insights, preventing Large Language Model (LLM) context drift over long execution spans.213214---215216## Configuration217218The system defaults to DeepSeek endpoints and supports any standard OpenAI-compatible API. LLM provider setup and all tunable runtime overrides ([ADR-017](docs/wiki/adr/017-config-via-api.md)) are documented in [docs/wiki/22-configuration.md](docs/wiki/22-configuration.md).219220---221222## Disclaimers & Backtest Limitations2232241. **No Financial Advice**: This system is a decision-support tool powered by LLMs. Output memos represent simulated reasoning based on deterministic data and do not constitute asset allocation advice.2252. **Backtest Time-Lock & Lookahead Guard**: The backtest engine (`scripts/backtest_runner.py`) has a hardcoded safety valve: **it rejects backtests for `decision_date > 2024-06-30` by default** (override with `--allow-lookahead`). Since mainstream foundation models have training cutoff dates around mid-2024, backtesting on later intervals introduces severe **Lookahead Bias** (model pre-training leakage). Parameter tuning, Optuna sweeps, and prompt optimization must run strictly on historical windows prior to June 30, 2024.226227---228229## Acknowledgments230231* [MiMo](https://mimo.mi.com/) — Special thanks to MiMo Quantitative Lab for sponsoring production-grade high-performance LLM inference (powering `mimo-v2.5-pro` long-horizon sweeps).232* [OpenClaw Dreaming Guide](https://dev.to/czmilo/openclaw-dreaming-guide-2026-background-memory-consolidation-for-ai-agents-585e) — Theoretical foundation for the three-phase sleep-cycle memory distillation framework.233234---235236## License237238This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.239
Full transparency — inspect the skill content before installing.