Frequently asked questions
AI agent marketplace FAQ
Everything teams ask before they deploy a ready-made AI agent. Straight answers on safety, no-code, integrations, lock-in, who builds the agents, and how creators get paid.
An AI agent marketplace is a store where you discover, buy, and deploy ready-made AI agents instead of building them from scratch. Agentmarketplace is the app store for AI agents: you browse vetted, security-scanned agents by category and use case, deploy the one you want in a single click onto your own stack, and run it on any model. Creators publish and monetize their agents, and buyers avoid lock-in because agents run on your tools, not trapped inside one platform.
Yes, when the agent has scoped, revocable permissions and a human stays in the loop. Every agent on Agentmarketplace is vetted and security-scanned before it is listed. When you deploy, you grant only the access the agent needs, you can revoke it instantly, and the agent keeps a human in the loop by default for anything sensitive. You also get an audit log of what the agent did.
No. Deploying an agent is one click. You pick an agent, connect the tools it needs through guided, no-code integrations, and it starts working. Developers can go deeper if they want, but the default experience is no-code, so anyone on the team can deploy and manage an agent.
It means agents are not trapped inside our platform or tied to one model vendor. Agents you deploy run on your own stack, connect to the tools you already use, and run on whatever model you choose. You can change the model, export your setup, or remove an agent at any time. We do not own your runtime, so you are never locked in.
Agents connect to the tools you already use, including Slack, HubSpot, Salesforce, Notion, Stripe, Gmail, GitHub, Shopify, Zendesk, Google Calendar, and Zapier, among others. Each agent lists the integrations it needs, and you grant access with scoped, revocable permissions when you deploy it.
Agents are built by vetted creators: independent AI builders, automation agencies, and software teams who publish their agents on Agentmarketplace and earn from sales. Every agent is reviewed and security-scanned before listing, and each listing spells out what the agent does, which tools it connects to, and the permissions it needs. You always know what an agent does before you deploy it.
Yes. Plans are month to month or yearly, and you can cancel any time. Because agents run on your own stack with scoped permissions, removing an agent or canceling your plan is clean and immediate, with no data held hostage.
No. Your data stays yours. Agents run with scoped access to the tools you connect, and your data is never used to train models, ours or anyone else's. You control what each agent can access and can revoke it at any time.
Creators publish their agent to the marketplace, set a price, and earn about 80% of every sale (roughly a 20% marketplace fee, the industry norm). Agentmarketplace handles discovery, billing, deployment, and trust signals like review and security scanning, so creators focus on building a great agent and get paid when teams deploy it.
Still have a question? Get in touch, or read how it works.