Via bankrate.com
The investing platform lets users build, deploy, and share automated trading agents using plain English prompts across stocks, ETFs, options, crypto, and bonds.
Public.com just opened an AI agent marketplace where users can browse, deploy, and customize automated trading agents for their portfolios, no waitlist required. The platform, which originally began rolling out AI agents to select users on March 31, 2026, is now making the feature broadly accessible along with something new: a marketplace where agents can be shared across the user base.
The move effectively turns Public into what the company calls the world’s first “agentic brokerage.”
How the agents actually work #
The core mechanic is simple enough. Users type what they want in natural language: something like “buy $500 of this ETF every time it drops 3% from its weekly high” or “rebalance my crypto allocation monthly.” The agent then monitors real-time market data and acts on those conditions.
Supported asset classes span stocks, ETFs, options, crypto, bonds, and cash management workflows.
The safety architecture is designed to keep humans in the loop. Every agent comes with an activity feed showing exactly what it’s doing and why. Users must approve actions before execution, and agents can be d or edited at any time.
The marketplace layer adds a social dimension. Users can now share their agent configurations with others, meaning someone who builds a clever dividend-harvesting strategy can make it available for the broader community to deploy. The platform does not currently support monetization of shared agents.
From click-based trading to intent-driven automation #
Public has been building toward this moment for a while. The company previously introduced AI-powered tools for custom index creation, letting users define thematic baskets of stocks using conversational prompts. AI agents represent the logical next step: moving from “help me pick what to buy” to “go ahead and buy it for me, under these conditions.” One open question is how regulators will view the proliferation of shared trading agents. If thousands of users deploy the same agent simultaneously, the resulting correlated trades could introduce new forms of market risk. Public’s current design, where each agent operates within individual accounts based on user-defined parameters, provides some insulation.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our