Integrations
Connects to the OpenAI API for hosted models and to Ollama through a custom base URL for local inference. Works with any Prisma project that has a schema file.
Learn morePrisma Assistant
Chat with your Prisma schema to get answers and generated queries.
Intermediate
Own Prisma Assistant? Verify ownership to get a dofollow link
What is Prisma Assistant?
Prisma Assistant is a developer tool that chats with your Prisma schema to get answers and generated queries. A local command line tool that opens a chat UI over your Prisma schema, answering questions and drafting TypeScript Prisma Client queries.
Prisma Assistant is a command line launcher that opens a chat interface over your existing Prisma schema. It runs a local server from the root of a Prisma project and serves a browser UI where you ask questions in plain language, such as listing users together with their related portfolios, and receive TypeScript Prisma Client code in response. The pitch is practical: skip the loop of opening the schema file, re-reading relations, and hand-writing a query that may or may not respect how the models connect. The setup is deliberately minimal. You export an OpenAI API key as an environment variable, run npx prisma-assistant@latest inside the project, and the tool prints a local URL to open. Two optional variables let you swap the model and override the API base URL, which is what makes the Ollama path work: point the base URL at a local Ollama endpoint, set the key to a placeholder string, and pick a model such as llama2. In that configuration the schema and prompts stay on your own hardware rather than going to a hosted provider. It fits best in the early stages of feature work, when a developer is new to a schema or wants a first draft of a query with includes and relations already wired up. The output is a starting point to review and adjust, not a replacement for understanding the data model. Because the assistant reads the schema that sits in the working directory, it is only as useful as that schema is complete and current. The project labels itself early alpha, so expectations should match. There is no pricing page, no account system, and no stated support channel. Security is framed around locality: the server runs on your machine and the maintainers say they do not hold your API key. Compared with general coding chat tools, the value here is narrower but sharper, since the conversation is grounded in one schema rather than an entire repository.
How do you use Prisma Assistant?
1Get an OpenAI API key
Create a key in the OpenAI API portal. If you plan to use a local model through Ollama instead, you can skip this and set a placeholder key later.2Set the key as an environment variable
On MacOS or Linux run export PRISMA_ASSISTANT_OPENAI_API_KEY=your_key in a terminal. On Windows use the set command. Set it permanently in your OS so it survives new sessions.3Launch the assistant from the project root
Open a terminal at the root of your Prisma project and run npx prisma-assistant@latest. The tool starts a local server and prints a URL.
4Configure a model or local endpoint
Optionally set PRISMA_ASSISTANT_OPENAI_MODEL and PRISMA_ASSISTANT_OPENAI_API_BASE_URL. For Ollama, point the base URL at http://localhost:11434/v1 and choose a model such as llama2.
5Chat with your schema
Open the printed URL in a browser and ask questions in plain language, for example listing users with their portfolios. Review the returned Prisma Client code before using it.
Pros and cons
Pros
- Generates Prisma Client code with relations and includes already mapped to your schemaAI
- Starts with a single npx command from the project root, no install or configuration fileAI
- Works with Ollama via a custom base URL, so a fully local model is possibleAI
- Model and API base URL are configurable through environment variablesAI
- Server runs on your own machine, so the schema is not uploaded by the tool itselfAI
Cons
- Explicitly labeled early alpha, so behavior and output can change without noticeAI
- Requires an OpenAI API key or a local model setup before anything worksAI
- No pricing, support or documentation pages beyond the single landing pageAI
- Missing an API key or misconfigured variable blocks the server with no in-app guidanceAI
- Only understands the schema in the current directory, not broader codebase contextAI
How much does Prisma Assistant cost?
Features
Conversational interface over an existing Prisma schema that answers questions and drafts TypeScript Prisma Client code, including relations and nested includes. Optional environment variables choose the model and override the API base URL, enabling a local Ollama setup. Runs as a local server started with npx from the project root.
Learn moreFrequently asked questions about Prisma Assistant
How do you use Prisma Assistant?
The walkthrough on this page covers 5 steps: 1. Get an OpenAI API key 2. Set the key as an environment variable 3. Launch the assistant from the project root 4. Configure a model or local endpoint 5. Chat with your schema.What platforms does Prisma Assistant support?
Prisma Assistant is available on Web App, MacOS, Windows and Linux.What does Prisma Assistant integrate with?
Connects to the OpenAI API for hosted models and to Ollama through a custom base URL for local inference. Works with any Prisma project that has a schema file.What are the limitations of Prisma Assistant?
Explicitly labeled early alpha, so behavior and output can change without notice. Requires an OpenAI API key or a local model setup before anything works. No pricing, support or documentation pages beyond the single landing page.
Status
StatusActive
Views0
Outbound clicks0
Added10/22/2026
Platforms
Web AppMacOSWindowsLinux
Pricing
Free
