AI Hedge Fund
Coordinate a desk of specialist quant prompts that hypothesize, backtest and rank crypto strategies, and reject the fragile ones.
- What
- Coordinate a desk of specialist quant prompts that hypothesize, backtest and rank crypto strategies, and reject the fragile ones.
- Cost
- Free
- Needs
- Use "AI Hedge Fund" with your Muse.
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
⚠️ **Trading warning / Avertissement trading** : informational only, not investment advice. This is a research desk, not a money manager: it produces hypotheses, backtests and reports, and a strategy it ranks highly can still fail forward and lose money if traded. Curated by Skill Harbor: the coordinator skill of DaviddTech's ai-trading-agent repo, a short manager prompt that runs an AI quant research workflow end to end. It assigns the work to specialist roles (quant mathematician for brand new strategies, mean reversion engineer, strategy optimizer, position optimizer, plus a risk manager whose job is to reject fragile, overfit or reckless systems, and a report writer) and drives a fixed loop: generate hypotheses, convert them to Pine Script, backtest with the Trader Dev MCP tools, validate across symbols and timeframes, rank by risk-adjusted quality, and prepare candidates for incubation or forward testing. Its operating rules are the honest core of the piece: never trust one backtest, never optimise before a baseline exists, never confuse leverage with edge, never ignore max drawdown, never use martingale without strict caps, never hide failed tests, and never claim production readiness without forward testing. Each cycle ends with a structured report (hypothesis, backtest matrix, best and worst results, robustness and risk scores, verdict, next action). From the DaviddTech/ai-trading-agent repository (MIT). Honest caveats: the skill coordinates the four other DaviddTech specialist skills in this catalogue, so it delivers its full value with them installed and with access to the third-party Trader Dev backtesting platform; on its own it is a disciplined workflow prompt, not software, and its rankings are only as good as the backtests fed into them. Skill Harbor never reviews the code, review it yourself before use.
Version:
Install
Copy the install package below, then paste it into MuseThe install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Use "AI Hedge Fund" with your Muse. Prerequisites: access to the Trader Dev MCP backtesting platform (a third-party service; check its current terms) and, for the full desk, the four companion DaviddTech skills (Quant Mathematician, Mean Reversion Engineer, Strategy Optimizer, Position Optimizer). This desk does research and backtesting only; it places no orders and needs no brokerage credentials. 1. Open the skill: https://github.com/DaviddTech/ai-trading-agent/blob/main/skills/ai-hedge-fund/SKILL.md and copy the full SKILL.md text. 2. Paste it into a chat with Muse and add: "Run one research cycle: pick a market universe, generate strategy hypotheses, and route each one to the right specialist." 3. Demand the full cycle report at the end, including the failed tests and the worst result, and treat incubation plus forward testing as mandatory before any strategy is called promising. Tip: the risk manager role is the valuable seat at this desk; if a run comes back with no rejections, the desk is not doing its job. Safety: a skill is plain-text instructions; it runs nothing by itself. Research only, no live trading. Informational only, not investment advice.
Saved to your recent installs. Find it anytime on /connect.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.