Build your own ChatGPT.

# Build an original app inspired by ChatGPT

You are an experienced product designer and full-stack engineer. Build a working, polished MVP from the brief below. Implement the core journeys end to end; a static mockup is not sufficient. Inspect the current environment, make sensible implementation decisions, and get started. Ask a question only if a missing answer genuinely blocks implementation.

Starting platform: Web. Explicit customized Product & audience and Tech stack requirements take precedence over this starting target.

## Product & audience
Build an original LLM chat wrapper called Thoughtdesk, inspired by the conversation workflow of ChatGPT. It is for users who want a focused chat workspace around a configured model provider. The main journey is creating a conversation, sending a message, receiving a real streamed response, stopping generation, and reopening history. This wraps a provider; it does not train or recreate a foundation model.

## Design
Use an original conversation workspace with a history sidebar, readable message content, a composer, and clear model/context status. Distinguish sending, streaming, interrupted, completed, and failed responses. Show tool or file features only if actually implemented; make copying and keyboard submission accessible.

## Features
1. Conversations: create/rename/archive/delete chats, search your own history, and persist message ordering. Keep each user’s conversations private and provide Markdown export of a chosen chat.

2. Real generation: call a configured provider from the server, stream output to the client, let users stop the request, and persist the final or partial result with its state. Never use hardcoded responses as a successful AI call.

3. Retry and context: retry a failed assistant response without duplicating user messages, record a bounded context policy, and warn when a conversation needs truncation. Preserve system instructions separately from user text.

4. Output rendering: safely render Markdown/code, provide copy controls, and show provider errors without exposing keys or private request logs. Do not execute model-generated code in the chat page.

5. Usage controls: configure an allowlist of available models, per-user request/concurrency limits, maximum input/output sizes, and provider-reported usage when supplied. Show setup-required state if credentials or a provider are unavailable.

## Tech stack
Use TypeScript for the browser and an authenticated server API, PostgreSQL for history, and the selected provider’s current official SDK/API. Keep API keys in server-side environment variables. Use an abortable streaming transport and explicit durable message states. Verify the provider’s current endpoint, model availability, and usage fields during implementation rather than pinning stale assumptions.

## Data & behavior
User; Conversation {ownerId, title, archivedAt}; Message {conversationId, role, content, parentId, generationId, state, createdAt}; Generation {providerRequestId, model, state, usage, errorCode}; UsageWindow; ModelConfiguration. Keep ownership checks on every message/history endpoint and redact credentials from errors.

Acceptance checks and scope:
Test a real streamed response, stop mid-generation, reload partial history, provider timeout/rate limit, duplicate retry, unauthorized conversation access, malicious Markdown, and missing credentials. Exclude model training, internet browsing, arbitrary tools, file retrieval, and unattended agents unless separately implemented.

## Originality and assets
The named product is a reference for general workflows only. Use an original product name, visual identity, wording, layout, and assets. Do not copy proprietary code, branded logos, distinctive screens, private APIs, paid templates, or protected media. Use original or appropriately licensed sample data and assets. Build from this brief rather than scraping the reference service. Do not imply affiliation with the referenced company.

## Implementation quality
- Make every visible control work, with clear loading, success, empty, validation, and failure states. Remove controls for unimplemented features.
- Support the target platform and responsive or resizable layouts where appropriate. Use platform-appropriate accessibility semantics, labels, keyboard operation, visible focus, sufficient contrast, and reduced-motion preferences. Provide alternatives for drag-only interactions.
- Validate inputs at the relevant boundary. Treat imported or user-generated content as untrusted; render it safely. Confirm destructive actions or offer undo.
- Keep state changes reliable across navigation and reloads. Never silently lose edits. Handle persistence and network errors with an actionable message.
- Keep the scope realistic. Prioritize the complete primary journey. If a custom Features section differs from the baseline, update the data model, design, and acceptance checks to match it; explicit custom requirements take precedence over baseline assumptions.
- Provide realistic original sample content and a useful empty state. Keep sample data clearly separate from user-created data.

## Delivery and verification
1. Inspect the repository and outline the implementation in a few sentences.
2. Implement the app and connect its actual data flow. Continue until the primary journey works.
3. Run appropriate checks and test the acceptance cases above on the target platform. Verify responsive layouts and keyboard navigation when browser tools are available. For native projects, build with the available SDK and test an emulator or device when possible; report any missing native verification honestly. For games, verify the rules and gameplay loop, not only the menus.
4. Fix observed failures, then provide the runnable result, setup/run instructions, a brief explanation of what is implemented, and any specific configuration or verification limits.

This is a focused MVP. Do not add billing, AI services, or unrelated features unless explicitly requested in the brief. Start building now.