Ajently

AI agent marketplace to discover, customize, and pay for agents either with fiat or crypto.

Ajently

Created At

Open Agents

Project Description

Ajently is an AI agent marketplace that unifies fragmented AI tools into one seamless platform. Instead of switching between multiple apps for writing, coding, tutoring, or content generation, users can discover and interact with specialized agents in a single environment.

Each agent is purpose-built for a specific task, with its own identity, prompt logic, and capabilities. Users can explore agents, understand what they do, and instantly use them through a focused chat interface. Beyond usage, Ajently allows users to customize agents by modifying prompts, defining behavior, and attaching knowledge.

On the supply side, creators can package and list their own agents, turning them into monetizable products. Ajently introduces a credit-based system powered by fiat payments, enabling users to pay for premium agent usage while giving creators a path to earn. This creates a complete ecosystem where AI tools are not just used, but discovered, customized, and monetized.

How it's Made

Ajently was built as a full-stack application using Next.js (App Router) and TypeScript, with Tailwind CSS for a clean and responsive UI. The architecture is designed to support a marketplace experience, real-time agent interaction, and a credit-based monetization system.

On the frontend, dynamic routes power key experiences such as agent exploration, detail pages, chat interfaces, and customization flows. The chat system supports markdown rendering, structured responses, and interaction controls like copy, share, and feedback.

The backend is handled through API routes within Next.js, managing authentication, agent execution, credit transactions, and payment confirmation. SQLite (via sql.js) is used as a lightweight database to store users, agents, and transaction data.

AI capabilities are powered through models from OpenRouter and 0G Compute, allowing access to multiple models for generating agent responses. One notable implementation detail is how agent configurations (identity, prompts, and attached knowledge) are dynamically composed at runtime to create highly contextual outputs.

For monetization, Ajently integrates Paystack and Interswitch Web Checkout for fiat payments. A server-side requery system ensures that credits are only issued after verified transaction confirmation, preventing client-side manipulation. This secure payment flow is tightly coupled with the credit system, making premium agent usage directly dependent on verified balances.

background image mobile

Join the mailing list

Get the latest news and updates

Ajently | ETHGlobal