ChatStack uses a team of specialised AI agents, powered by OpenAI, that work together in sequence. You describe your project idea in the chat and the agents guide you through each step: interviewing you for clarity, writing user stories, creating detailed requirements and specifications, and producing a cost estimate.

The entire process happens through a conversational interface — just type naturally and the agents do the rest. The final output is a comprehensive PRD in structured data format which is AI-readable by coding agents such as Cursor or Claude Code

The process takes between 10–30 minutes depending on the complexity of your project and which mode you select: "Quick Estimate", "Normal", or "Detailed".

It may seem slow, but consider how long it would take a human to produce the same output. Each agent is specially trained to complete their complex task based on specific rules and their knowledge base. To ensure the output is structured data (JSON), they need to continuously shape their output.

The wait is worth it: the output is a comprehensive PRD in structured data format which is AI-readable by coding agents such as Cursor or Claude Code. This saves significant time downstream by enabling multi-phase planning, avoiding context-drift, and generating clean code.

Yes. Here is a summary of our core promises:

1. You Own Your Ideas: We do not claim ownership of the business concepts or prompts you enter. You own the output the AI generates for you.
2. No Training on Your Data: Your inputs are NOT used to train the AI models. Your ideas will not help your competitors.
3. Strict Confidentiality: We treat your data as Confidential Information. We will not share it with third parties unless strictly necessary to operate the service (e.g., sending the text to the AI engine to generate an answer).
4. Security First: Your data is stored on secure servers in the United Kingdom. We use industry-standard encryption to protect it.
5. Your Responsibility: AI can make mistakes. Please verify any facts, figures, or code generated by the tool before you build your business on them.

Read the full terms

When the AI agents finish, you will be asked to provide your name, email address and country. You will then receive an email containing your complete Product Requirements Document.

The PRD is delivered as an email summary, ready to use with AI coding tools like Cursor or Claude Code. The email will include a link to our Product Management Portal where you can review, improve, design and plan your project before generating the Build prompt.

You can also book a call with the team to discuss next steps.

In marketing we sometimes cite an approximate 65% gap between ChatStack's indicative build estimates and the benchmark ranges we use for comparison. Those benchmarks are built from typical market pricing assumptions for comparable software scope (for example, blended day rates and effort ranges for similar-sized projects), not from a single vendor quote or a guarantee of what you will pay.

The percentage is a representative illustration across common project shapes; your outcome can be higher or lower depending on geography, team seniority, risk buffers, what is in or out of scope, and how vendors price. Treat ChatStack's estimate as planning guidance, then validate with real proposals.

For how pricing works on the site, see how much does it cost to build an app.

ChatStack is designed for any individual or team looking to launch a new product, whether you are a start-up, a development company, or an enterprise. It is particularly valuable for:

• Building complex solutions.
• Solving the unique problems of large teams, such as ensuring thorough documentation, signoff, and adherence to strict coding standards.

Currently ChatStack is completely free. We (App Developer Studio) use it as a sales tool to find new clients. Once you've generated your PRD and reviewed the cost estimate, we'll typically set up a call to discuss how you'd like to proceed: we can design and develop the entire project, or we can work with your development team to help them integrate the PRD output with their preferred technology stack.

No. ChatStack uses a proprietary agentic workflow and RAG vector stores specific to software architecture and project management logic to ensure technical accuracy that generic LLMs cannot provide.

Yes. ChatStack can be used to document new features or “reverse-engineer” requirements for existing codebases to help AI tools understand the current system architecture. Let's set up a call to discuss how we can help you.

ChatStack provides an MCP server interface, allowing AI agents (like Claude Desktop or Cursor) to “read” your PRD data in real-time as they write code.