AI strategy building
Describe a goal, and the product generates systematic strategy candidates. The homepage shows examples such as long-term growth, drawdown protection, trend following, and dip buying.
Portfolio Lab
Portfolio Lab is an AI-assisted strategy builder and validation platform for self-directed investors. It lets users create systematic trading rules, test them in live paper markets, and have Claude, ChatGPT, or another MCP-compatible agent mirror the plan in their brokerage account.
Strategy research and agentic execution for self-directed investors
Portfolio Lab is a strategy research and simulation product for investors who want AI-assisted trading rules without building the full system themselves. It helps users construct systematic investment strategies, validate them in live paper markets, and then let an AI agent follow those rules inside the user’s own brokerage account.
The site positions the product as the “strategy layer” for agentic trading: Portfolio Lab defines what to hold, how much to hold, and when to move into cash or a hedge, while Claude, ChatGPT, or another MCP-compatible agent handles execution through the brokerage workflow. The disclosures also state that the product is for education and simulation, not investment advice.
Core capabilities
Live validation in paper portfolios
Strategies can be tested in live paper trading with real market data and fills, letting users see how a strategy behaves before committing capital.
Agent-driven execution
The product can move a validated strategy into an agentic workflow where Claude, ChatGPT, or another MCP-compatible agent reads the daily plan and mirrors it in a brokerage account.
Visible strategy mechanics
The system surfaces target holdings, trade timing, and the stated reason for a move to cash or a hedge, so the logic behind a strategy is visible rather than hidden.
Multiple strategy styles
The site lists multiple strategy types, including tactical long-only, tactical long/short, momentum, mean reversion, and hedged variants with long, cash, hedge, or inverse exposures.
Multi-source model inputs
The disclosures say the system uses AI and machine-learning models across fundamental, technical, macroeconomic, and alternative data.
Common use cases
Create a systematic portfolio from a goal
For investors who want a rules-based portfolio but do not want to design the entire system manually, Portfolio Lab can generate multiple strategy candidates from a stated goal and show the holdings behind each one.
Validate a strategy before funding it
For users who want to test an idea before using real capital, the product supports live paper validation with market data and fills, allowing the strategy to be assessed in a simulated environment.
Delegate execution to an agent
For people who want an AI assistant to execute a strategy without improvising, Portfolio Lab provides the daily plan that Claude, ChatGPT, or another MCP agent can mirror in the brokerage account.
Build defensive or risk-aware strategies
For investors who prefer defined downside controls, the product includes strategy types that shift into cash, add a hedge, or switch to an inverse exposure when risk rises.
Compare different strategy styles
For users comparing systematic approaches, the site offers tactical long-only, long/short, momentum, and mean reversion rule sets, each with different asset counts and exposure styles.
Pros and Cons
Pros
- Supports both strategy design and live paper validation in one workflow.
- Makes the strategy logic visible, including holdings, trade timing, and reasons for cash or hedge shifts.
- Works with Claude, ChatGPT, and other MCP-compatible agents for execution.
- Offers several systematic strategy styles rather than a single preset portfolio.
- Uses market-data-driven models and discloses the inputs and limits of the process.
Cons
- Pricing details are limited on the source pages, and the pricing URL shown in the audit returns a not-found page.
- The product is described as educational and simulation-oriented, so it does not remove trading risk or guarantee results.
- The agentic workflow depends on an external agent and brokerage setup, so users still need to connect and manage those tools separately.
FAQ
Does Portfolio Lab place trades directly?
No. Portfolio Lab says it serves your strategy's daily plan through a read-only connection. Your agent places the trades through your brokerage's own tools, on your instructions.
What does my agent see from Portfolio Lab?
It provides today's target allocation for your portfolio, including tickers and percentages. The disclosures say it does not expose brokerage credentials, balances, or personal data.
Which agents and brokers work with it?
Claude and ChatGPT are supported today, along with any client that supports MCP. The FAQ also mentions Robinhood Agentic Trading as a brokerage path, and says any broker your agent can trade with can work.
What does it cost to use the agentic workflow?
Building and proving strategies is free. The FAQ says agentic trading is part of the Analyst plan, while paper strategy building and validation are available without charge.
Is this investment advice?
No. Portfolio Lab is presented as a research and simulation tool, and the disclosures state that it is not investment advice. Any trade your agent places is your decision and responsibility.
Quick Facts
- Category
- AI investing platform
- Primary workflow
- Build a strategy, validate it in paper trading, then let an agent mirror it in your brokerage account
- Supported agents
- Claude, ChatGPT, and any MCP-compatible client
- Company
- alphaAI Capital Management LLC
- Site
- portfoliolab.ai
- Regulatory note
- Disclosures describe alphaAI Capital as an SEC-registered investment adviser; Portfolio Lab itself is presented as a research and simulation tool
Portfolio Lab Alternatives
Lasso
Lasso is an ecommerce product data platform for enriching catalog records, processing supplier files, generating product content, and tracking competitors. It includes a web app, REST API, SDK, and MCP server.
Botacts
Botacts is a web directory for finding AI bots and agents you can contact by phone, email, SMS, WhatsApp, Telegram, or Signal.
Tavus
Tavus is an AI video platform for real-time face-to-face agents, digital twins, and AI companions, with APIs and multilingual workflows for developers.
HiringPartner.ai
HiringPartner.ai is an autonomous AI recruiting platform for sourcing, screening, and interviewing candidates 24/7 with ATS-connected workflows and bulk resume uploads.
Macuse
Macuse is a macOS app that lets AI assistants control native Mac apps through MCP and use Computer Use for any app, with on-device automation.
AgentMail
AgentMail is an email inbox API for AI agents to create, send, receive, and search messages via REST APIs and SDKs. Built for threaded workflows, verification, support, scheduling, and approvals.