Robinhood gave its AI trading agent a new skill this week. The agent can now run technical indicators on demand. Users prompt it to calculate RSI, MACD, Bollinger Bands, and roughly 15 more signals. They pick the stock and the timeframe themselves. Notably, the feature went live to all US users at once, rather than a limited beta.
This upgrade closes an obvious gap in Robinhood’s AI strategy. Previously, the agent could place trades but could not analyze the charts behind them. Now it does both. As a result, a single conversation can move from analysis to execution.
Your agent speaks technicals now.
— Robinhood (@RobinhoodApp) July 30, 2026
Ask it to run RSI, MACD, Bollinger Bands, and 15 more indicators on stocks and timeframes you choose. Live for everyone in the US today. pic.twitter.com/871RYNqE8M
What Robinhood Actually Shipped
The new capability sits on top of Robinhood’s Agentic Trading system. Instead of static charts, the agent computes indicators live when asked. For example, a user might request the 14-day RSI on a stock across a four-hour timeframe. The agent returns the reading and can explain what it suggests. Additionally, it can layer multiple indicators into one view.
The indicator set covers the standard categories traders rely on. RSI and MACD measure momentum. Bollinger Bands track volatility. Moving averages follow trend direction. Together, roughly 18 indicators give the agent a working vocabulary for technical analysis. Importantly, users control the parameters rather than accepting fixed defaults.
How It Connects to Agentic Trading and MCP
Robinhood built this on a foundation it laid in May 2026. On May 27, the company published a post titled “Robinhood is Now Open to Agents.” That launch introduced Agentic Trading, which lets outside AI models act on a user’s behalf. The plumbing is the Model Context Protocol, an open standard for connecting agents to real tools.
Through MCP, users link agents like Claude, ChatGPT, or Gemini to Robinhood directly. The agent then reads portfolio data and submits equity orders. However, it works only inside a separate, pre-funded account. This sandbox keeps any losses contained to money the user deliberately set aside. Robinhood remains the first mainstream US retail brokerage to open live trading this way.
The Analysis Layer: Cortex and Legend
Robinhood already offered technical tools before this release. Its Cortex assistant lets users build custom scans through plain-language prompts. For instance, someone can ask for stocks where RSI drops below 30 while price holds above the 200-day average. Meanwhile, the Legend platform carries more than 80 charting indicators for active traders.
The new agent feature borrows that analytical depth and hands it to the autonomous layer. In contrast to Cortex, which surfaces ideas for a human to act on, the agent can research and trade in one flow. This blurs the line between assistant and operator. Consequently, the technical vocabulary now lives inside the tool that also executes.
Adoption and the Numbers Behind It
Robinhood has real traction to point to here. More than 100,000 users have opened funded Agentic Trading accounts since May. The broader eligible base sits near 27.5 million customers. That gap shows how much room remains for the feature to spread.
The company’s finances give it fuel to keep building. In its second-quarter 2026 results, Robinhood reported $1.31 billion in revenue. That figure climbed from $989 million a year earlier. Earnings reached 62 cents per share, well above the 42 cents analysts expected. Strong results let the firm keep investing in AI infrastructure.
The Reliability Question
Robinhood has not oversold what these agents can do. CEO Vlad Tenev has openly flagged their limits. He notes that the models “sometimes fight you” when asked to trade. Users occasionally work hard to get an agent to follow instructions. Additionally, he admits the current setup carries friction, since people must stitch a separate AI app to their brokerage.
These caveats explain the cautious design. The sandbox account, the equities-only scope, and the funded-money limit all reflect distrust of full autonomy. For readers, the takeaway is balanced. The agent can now read charts and act, yet it still needs a human watching closely.
Why It Matters
This release pushes retail trading toward a conversational model. Traders once jumped between charting apps, screeners, and order tickets. Now a single agent can handle the whole loop. That convenience is real, and it lowers the skill barrier for technical analysis.
Still, the same convenience raises the stakes. An agent that misreads a signal can now act on that mistake quickly. Therefore, Robinhood’s guardrails matter as much as its features. The company is betting that sandboxed autonomy earns trust before full autonomy arrives.
*Disclaimer: News content provided by Genfinity is intended solely for informational purposes. While we strive to deliver accurate and up-to-date information, we do not offer financial or legal advice of any kind. Readers are encouraged to conduct their own research and consult with qualified professionals before making any financial or legal decisions. Genfinity disclaims any responsibility for actions taken based on the information presented in our articles. Our commitment is to share knowledge, foster discussion, and contribute to a better understanding of the topics covered in our articles. We advise our readers to exercise caution and diligence when seeking information or making decisions based on the content we provide.
