General Overview of our App

InspiratiAI agents are creating new ways to research and trade prediction markets, but getting started requires technical knowledge and tools that many people don’t have. We wanted to make agentic trading more accessible while giving users clear control over what an agent can do with their money on

What it does

We learned that building a trading agent involves more than connecting an AI model to an account. Users need understandable limits, approval requirements, and visibility into every decision. Governance must operate independently of the agent so the agent cannot change or bypass its own restrictions.

How we built it

We collaborated through GitHub branches and pull requests, using Cursor and AI coding assistants like Claude. Python powers the backend, while HTML, CSS, and JavaScript power the interface. We integrated Kalshi for prediction market connectivity and Auth0 for individual sign-in, then deployed the app on Render. Our simulated demo lets users explore the workflow without real funds or API keys.

Challenges we ran into

Our biggest challenge was translating plain-language trading requests into actions that consistently respect user-defined rules. We separated each agent’s mandate from account-wide safety controls and distinguished trades that can execute automatically, require approval, or must be blocked. We also worked through credential protection, personal workspace isolation, and keeping the deployed website aligned with our latest code.

Accomplishments that we're proud of

We built a working prototype that connects AI trading agents with independent governance. Users can set agent-specific mandates and account-wide limits, review trades requiring approval, and see why a proposal was allowed or blocked. We also created a simulated demo and deployed the app so others can explore the experience.

What we learned

We learned that making AI trading accessible means making its boundaries understandable. An agent needs clear permissions, and those permissions must be enforced independently of the model. Collaborating through GitHub also taught us how to coordinate changes, review each other’s work, and troubleshoot deployment issues as a team.

What's next for GovernTradeAI

We plan to expand beyond Kalshi with Polymarket connectivity, add scheduled trading across multiple days, and explore a preconfigured agent that lowers the setup barrier for new users. We also want to improve trade monitoring, strengthen testing, and gather user feedback to make governance decisions easier to understand.