Prompt-based agent builder
Create an agent by writing plain-language instructions, selecting a model, connecting knowledge, previewing behavior, comparing models, and running evaluations before publishing.
Kopai
Freemium
Kopai is a hosted platform for building, running, publishing, and monetizing domain-specific AI agents. It supports internal team agents, marketplace listings, and deployment to a creator’s own site through live APIs.
Kopaiとは？
Kopai is a serverless cloud and marketplace for domain-specialized AI agents. It lets creators build agents from prompts, connect documents or organizational data, test models, evaluate responses, and publish the result for private team use or public marketplace access. Kopai handles the underlying retrieval, orchestration, model routing, storage, streaming, infrastructure, and message billing.
An agent can stay inside an organization or be submitted for marketplace evaluation. Team workspaces provide role-based access, shared knowledge libraries, project separation, and usage and cost reporting. Public agents can charge a publisher-set markup, while the platform itemizes runtime and publisher charges. Kopai also offers live APIs for running an agent in another product or site; its SDK is described as forthcoming.
Kopaiでできること
Managed retrieval and infrastructure
Kopai operates retrieval, model routing, file storage, streaming, vector-database infrastructure, and provider credentials so teams do not need to deploy or maintain those components.
Evaluation and grounding checks
Responses can be checked against the sources they cite. Evaluation results are associated with the version that earned them, so editing a prompt requires another pass.
Organization workspaces
Invite team members with owner, admin, or developer permissions. Organizations can share a knowledge library across agents, isolate projects, and keep agents private by default.
Marketplace and billing ledger
Approved agents can be listed with a publisher markup. Kopai records message charges, runtime usage, payouts, and earnings in an itemized ledger, with creators receiving 70% of their markup.
Resumable API and conversations
Conversations can resume after a dropped connection, and live APIs allow an agent to be used in another product or on the creator’s own site. The site says an SDK is shipping next.
利用シーン
“Internal policy and knowledge assistants”
A company can attach handbooks, policies, or other internal documents to a private agent so employees can query shared organizational knowledge without making the agent public.
“Document and workflow automation”
Creators can build agents for tasks such as checking supplier invoices against payment terms, analyzing data, supporting workflows, or generating documents from connected knowledge.
“Product and service comparison research”
CompareCraft demonstrates a research workflow that gathers evidence from official documentation, GitHub, Reddit, and independent reviews, then presents comparisons as tables, charts, and sentiment breakdowns.
“Public expert tools”
A creator can evaluate a domain-specific agent, submit it to the marketplace, set a markup, and let users try its initial free messages before paid credit usage begins.
“Embedded agent experiences”
Teams or product builders can use Kopai’s live APIs to place an agent in their own product or website while keeping hosting, model access, billing, and measurement managed by Kopai.
よくある質問
What does Kopai manage for an agent creator?
Kopai says it manages retrieval, orchestration, model routing, file storage, streaming, infrastructure, provider keys, and billing. The creator supplies the instructions, knowledge, and intended use of the agent.
Can an agent be used only by my team?
Yes. Organization agents are private by default and do not need to be listed, priced, or evaluated for public marketplace access. Teams can invite members and assign owner, admin, or developer roles.
How are users charged?
Users spend credits per message. The charge reflects the runtime used plus any publisher markup, and both parts are itemized. Kopai offers monthly credit plans, one-time wallet top-ups, and auto-reload; unused monthly credits roll over.
What is required before publishing an agent?
Marketplace agents must pass Kopai’s evaluation before they can be listed. The site states that an evaluation pass belongs to the version that earned it, so changing the prompt requires the agent to be evaluated again.
Can Kopai agents be embedded elsewhere?
The site says live APIs support running an agent on a creator’s own domain or inside another product, including streaming chat and native agent APIs. An SDK is described as forthcoming.
クイック情報
- Product type
- Serverless cloud and marketplace for AI agents
- Primary users
- AI agent creators, product builders, and organizations
- Deployment options
- Private organization workspace, public marketplace, or own site via live APIs
- Team roles
- Owner, Admin, and Developer
- Billing model
- Credits per message, with runtime and publisher charges itemized
- Example data sources
- Postgres read-only replicas, Snowflake, SharePoint, Confluence, S3, and existing vector stores; setup may involve Kopai’s team
Kopaiの代替品
Gemini Enterprise Agent Platform
cloud.google.com
Gemini Enterprise Agent Platform, formerly Vertex AI, is Google Cloud’s platform for developers and technical teams to build, deploy, govern, and optimize AI agents, generative AI applications, and machine learning models.
Gemini Enterprise Agent Platform
cloud.google.com
Gemini Enterprise Agent Platform is Google Cloud’s platform for developers and technical teams to build, deploy, scale, govern, and optimize enterprise AI agents and generative AI applications using enterprise data.
Context
context.ai
Context は、顧客インフラ上で AI エージェントを構築・デプロイ・改善できる企業向けプラットフォームです。ワークスペース、ランタイム、コンテキスト、評価ツール、コネクタ、IdP ベースのアクセス制御を備えています。
OneReach.ai
onereach.ai
OneReach.ai GSXは、企業向けAIエージェントを大規模に構築・運用・統制する基盤です。
Octomind Cloud
octomind.run
Octomind Cloud runs AI agents on persistent cloud machines for research, writing, web work, software development, and scheduled routines. Users give instructions in plain language and can receive results through the web panel, Telegram, WhatsApp, or Slack.
Stagehand
www.stagehand.dev
Stagehand is an SDK for building browser agents with Playwright-style APIs, natural-language actions, structured extraction, and self-healing behavior. It supports TypeScript, Python, and Go for agents that need to operate web pages with browser-level control.