Learn Ethereum, Gatsby and Postgres

PostgreSQL is a powerful, open source object-relational database system.

postgres

We create a PostgreSQL join table for `likes` in the [Supabase](https://supabase.com) dashboard. This has a many-to-many relationship between the `profiles` and `tweets` table, allowing us to store each instance of a like between a user and a tweet.

Additionally, we create Row Level Security (RLS) polices to enable `select`, `insert` and `delete`.

## Code Snippets

**Create likes table**

```sql
create table public.likes (
  id uuid default gen_random_uuid() primary key,
  created_at timestamp with time zone default timezone('utc'::text, now()) not null,
  user_id uuid references public.profiles on delete cascade not null,
  tweet_id uuid references public.tweets on delete cascade not null
);
```

**Enable Row Level Security**

```sql
alter table public.likes enable row level security;
```

**Enable insert action with RLS policy**

```sql
create policy "authenticated users can insert their own likes" ON "public"."likes"
as permissive for select
to authenticated
using (user_id = auth.uid());
```

**Enable delete action with RLS policy**

```sql
create policy "authenticated users can delete their own likes" ON "public"."likes"
as permissive for delete
to authenticated
using (user_id = auth.uid());
```

**Enable select action with RLS policy**

```sql
create policy "authenticated users can select likes" ON "public"."likes"
as permissive for select
to authenticated
using (true);
```

## Resources

- [Managing user data docs](https://supabase.com/docs/guides/auth/managing-user-data)
- [Row Level Security docs](https://supabase.com/docs/guides/auth/row-level-security)

Create a PostgreSQL Join Table in Supabase Studio

[Supabase](https://supabase.com/) has an `auth.users` table that contains information about our user and their session. We want to display the user's name, username and avatar alongside their tweets, but the `auth.users` table cannot be publicly accessible, as it contains sensitive information.

In this lesson, we create a new table called `profiles` and populate it with the data we want to display from the `auth.users` table. Additionally, we set up a PostgreSQL Function and Trigger to create a new profile for any user added to the `auth.users` table.

Lastly, we create an RLS policy for the `profiles` table to enable read access, and re-generate our TypeScript definitions file to contain our new table.

## Code Snippets

**Create profiles table**

```sql
create table public.profiles (
  id uuid not null references auth.users on delete cascade primary key,
  name text not null,
  username text not null,
  avatar_url text not null
);
```

**Enable Row Level Security**

```sql
alter table public.profiles enable row level security;
```

**Enable read access with RLS policy**

```sql
create policy "anyone can select profiles" ON "public"."profiles"
as permissive for select
to public
using (true);
```

**Create PostgreSQL Function to create profile**

```sql
create function public.create_profile_for_user()
returns trigger
language plpgsql
security definer set search_path = public
as $$
begin
  insert into public.profiles (id, name, username, avatar_url)
  values (
    new.id,
    new.raw_user_meta_data->'name',
    new.raw_user_meta_data->'user_name',
    new.raw_user_meta_data->'avatar_url'
  );
  return new;
end;
$$;
```

**Create PostgreSQL Trigger to create profile**

```sql
create trigger on_auth_user_created
  after insert on auth.users
  for each row execute procedure public.create_profile_for_user();
```

## Resources

- [Managing user data docs](https://supabase.com/docs/guides/auth/managing-user-data)
- [Row Level Security docs](https://supabase.com/docs/guides/auth/row-level-security)

Automatically Generate a Profile for Every User with PostgreSQL Function Triggers

[Supabase](https://supabase.com/) uses the `auth.users` table to store information about our users and their sessions. In this lesson, we add a column to the `tweets` table to associate each tweet with one of our users.

Additionally, we set up cascading deletes, so when a user is deleted from the `auth.users` table, their tweets are also deleted.

Lastly, because our database schema has changed, our TypeScript definitions are now incorrect. Therefore, we use the Supabase CLI to introspect our PostgreSQL schema and create a new `database.types.ts` file.

## Code Snippets

**Add foreign key relationship to auth.users**

```sql
alter table public.tweets
add user_id uuid references auth.users on delete cascade not null;
```

**Generate TypeScript definitions**

```bash
npx supabase gen types typescript --project-id your-project-id > lib/database.types.ts
```

## Resources

- [Supabase Auth docs](https://supabase.com/docs/guides/auth)
- [Supabase CLI](https://supabase.com/docs/guides/cli)
- [Generating TypeScript definitions](https://supabase.com/docs/guides/api/rest/generating-types)

Setup a Foreign Key relationship between PostgreSQL tables

Next.js is a minimalistic framework for server-rendered React applications.

Since we're using [Supabase](https://supabase.com/) to authenticate our users, it knows who our user is throughout the rest of the Supabase environment.

In this lesson, we modify our Row Level Security (RLS) policy to only apply to the `authenticated` role to ensure that only authenticated users can select tweets.

Additionally, we use the `useRouter` hook and `router.refresh()` to run our Server Component again and fetch fresh data from Supabase, after the user signs in or out.

## Code Snippets

**Enable authenticated read access with RLS policy**

```sql
create policy "anyone can select tweets" ON "public"."tweets"
as permissive for select
to authenticated
using (true);
```

## Resources

- [Row Level Security docs](https://supabase.com/docs/guides/auth/row-level-security)
- [useRouter hook](https://nextjs.org/docs/app/api-reference/functions/use-router#userouter)

Restrict Access to Authenticated Users with Supabase RLS Policies

[Supabase](https://supabase.com/) is a collection of open source tools that wrap around a PostgreSQL database - Hosting, Auth, Realtime, File Storage, Edge Functions etc. In this lesson, we go to [database.new](https://database.new) to create a free Supabase project.

Additionally, we create a `tweets` table and populate it with some seed data.

## Code Snippets

**Create a `tweets` table**

```sql
create table if not exists tweets (
  id uuid default gen_random_uuid() primary key,
  created_at timestamp with time zone default timezone('utc'::text, now()) not null,
  title text not null
);
```

**Insert tweets**

```sql
insert into tweets(title)
values
  ('first tweet'),
  ('second tweet'),
  ('third tweet');
```

## Resources

- [Supabase docs](https://supabase.com/docs)
- [Supabase SQL Editor](https://app.supabase.com/project/_/sql)

Create a New Supabase Project and Basic PostgreSQL Schema

Ethereum is a decentralized, open-source blockchain with smart contract functionality. 

ethereum

Welcome to Noah's course on Introduction to Smart Contracts with Ethereum and Solidity. 

Introduction to Introduction to Smart Contracts with Ethereum and Solidity

Web3 is a paradigm shift in how applications are interacted with, ran, and built. 

It’s not just a Pump-and-Dump scheme on cryptocurrencies or Bored Ape NFT profile pictures you can put on Twitter.

A fundamental tenet of Web3 is bringing ownership back to the user. You no longer have to trust that your data is being properly handled on someone else’s server (although, currently Web3 ecosystem mainly depends on centralized infrastructure such as GitHub).  

This ownership is achieved through blockchain, cryptocurrencies, NFTs, and smart contracts.

A great example of this type of ownership is through how you exchange money in Web3. There is no intermediary like Stripe, or PayPal to process your transactions. You can even build your own smart contracts that receive crypto (e.g. Ethereum or ‘eth’ for short) and then process that eth how you’d like.

Building this type of smart contract is exactly what Matías is going to show you how to do right now.

Matías builds out a Web3 application that allows you to receive tips in the form of eth in a tip jar on your website. This means your application will be backed by a smart contract that’s accepts eth, stores data, and emits Solidity events.

The frontend application uses SvelteKit which you’ll see how easy it is to get up and be productive with in minutes. A huge benefit here is that the JavaScript you write is just that, JavaScript. You’ll be able to take what you learn and implement it into any framework of your choosing.

While you build this application you will learn how to:

- Write a smart contract using Solidity language
- Test the smart contract using just JavaScript through the waffle framework
- Communicate and interact with the smart contract using ethers.js in a SvelteKit application
- Deploy your smart contract to make it public and accessible to anyone

Solidity is an object-oriented, high-level language for implementing smart contracts.

solidity

Full Stack Web3 and Blockchain Development on Ethereum with SvelteKit

There are several Ethereum test networks like Ropsten, Rinkeby and Kovan that you can use to deploy your contract to have it publicly available without having to use the mainnet and real ETH.
In this course you'll deploy to Rinkeby netwok (but, the process is the same for the other networks).

The first step will be update your Metamask Wallet to connect to the Rinkeby network.

Next, you'll need some test/fake Ethere to use for testing the Dapp in the network, for that you'll need to use a [faucet like this one](https://www.rinkebyfaucet.com/). Just visit that faucet and add your wallet address.

You will also deploy your web application to vercel to see your project live on the internet!

Deploy your Smart Contract to Goerli Test Network and App to Vercel

The magical disappearing UI framework

 - The web's JavaScript bloat crisis, solved. Svelte turns your templates into tiny, framework-less vanilla JavaScript.
 - Simple and familiar. Build apps out of composable, easy-to-write blocks using languages you already know.
 - Super fast, rock solid. Compile-time static analysis ensures the browser does no more work than it needs to.

svelte

Now that we have an application that is accepting tips to the smart contract that was defined, we need to be able to withdraw the funds that are sent to the contract. The smart contract function for this is already defined so we will just need to call the function from the front end.

To do this we will create a new route to display the contract information. We will then check if the current user is the contract owner and if so allow that user to withdraw the contract funds by clicking a button that we wire up to call the withdraw function from the smart contract.

## Important

SvelteKit is still in beta. Expect bugs! Read more here, and track progress towards 1.0 here. 

This lesson was recorded before the latest breaking changes, the content is still valid but it sometimes references a file that does not exists anymore.

Now: 
`src/routes/index.svelte` is called `src/routes/+page.svelte`

`src/routes/__layout.svelte` is called `src/routes/+layout.svelte`


Everything else is 100% valid. This particular content will be updated when Sveltekit reaches 1.0.0 stable.

Withdraw eth from a Smart Contract in a Svelte Application

It is time to interact with the contract, let's send some eth!

For this you'll need a form that is rendered only when the wallet is connected.
This form use the `sendTip` function and gather some required information. On submit, it will read the `sendingTip` loading state variable and disable or enable the button.

## Important

SvelteKit is still in beta. Expect bugs! Read more here, and track progress towards 1.0 here. 

This lesson was recorded before the latest breaking changes, the content is still valid but it sometimes references a file that does not exists anymore.

Now: 
`src/routes/index.svelte` is called `src/routes/+page.svelte`

`src/routes/__layout.svelte` is called `src/routes/+layout.svelte`


Everything else is 100% valid. This particular content will be updated when Sveltekit reaches 1.0.0 stable.

Create a Form in Svelte and Send a Tip to an Solidity Smart Contract

Now is time to setup the client to talk with the contract, for that you will need to store and access the contract ABI (this is a file generated by the contract compilation) as well as use ethers to set up a connection to the contract. 

Once set up is complete, we will listen to any tip events that come in from the contract and display tip information in the application.

## Important

SvelteKit is still in beta. Expect bugs! Read more here, and track progress towards 1.0 here. 

This lesson was recorded before the latest breaking changes, the content is still valid but it sometimes references a file that does not exists anymore.

Now: 
`src/routes/index.svelte` is called `src/routes/+page.svelte`

`src/routes/__layout.svelte` is called `src/routes/+layout.svelte`


Everything else is 100% valid. This particular content will be updated when Sveltekit reaches 1.0.0 stable.

Interact with a Solidity Smart Contract Using Ethers and the Contract ABI in SvetleKit

Now the UI is capable to connect to the Ethereum network by using the wallet, but we need two more things to continue.

1. Avoid re-connecting every time the browser is refreshed
2. Get more information about the user and the network.

For the first, let's run a little function every time the component is mounted.

Then, let's read and display some more information:

- The user balance
- The network that the user is currently connected

## Important

SvelteKit is still in beta. Expect bugs! Read more here, and track progress towards 1.0 here. 

This lesson was recorded before the latest breaking changes, the content is still valid but it sometimes references a file that does not exists anymore.

Now: 
`src/routes/index.svelte` is called `src/routes/+page.svelte`

`src/routes/__layout.svelte` is called `src/routes/+layout.svelte`


Everything else is 100% valid. This particular content will be updated when Sveltekit reaches 1.0.0 stable.


Retrieve Ethereum Network and Account information on Mount and Display it in Svelte

The first step in this part of the journey, will be to setup a simple UI that allow you to connect the current user to the blockchain through it's wallet.

To do this, let's write some code.

Sveltekit applications works by using a file system routing method, meaning that you need to create files under the `src/routes/` folder and those will be then reflected as routes (urls) in your app. Let's keep ti simple and just use the default route here, open `src/routes/index.svelte` and delete the content of it.

You're first step will be create a button that allow the user to "sing-in" to your new dapp and allow the UI to react to that showing a different content if the authorization process was successful.


Connect a Metamask Account to a Svelte Application

Before jumping into the Dapp code you need to setup an Ethereum wallet.
There are bunch of wallet applications and types outher, but for this project we're going to use Metamask.

### What is an ethereum Wallet

A wallet is an application that let's you interact with your Ethereum account. You're wallet is the one that works as a gate between the Dapp and your data, through it the applications can read your balance, send transactions, etc.

You can swap wallet providers at any time or manage multiple accounts in one application. This is because the wallet are non-custodial, meaning that they don't hold your assets, you do.

Just to set some concepts down:

- An Ethereum account is an "entity" that can send transactions and store some assets, it "lives" inside the blockchain.
- The accout have an address (like your profile URL of some social app or your email address) that can be used to send or receive funds.
- A wallt is an application that allow you to interact with the Ethereum account.

There are different types of Wallets:

- Hardware wallets that hold your crypto assets offline - The most secure of all
- Mobile apps, like Argent.
- Web wallets, that let you interact through the browser: Like Metamask.
- Desktp apps

Set up a Metamask Wallet with your local Ethereum Network

Now you need a way to keep a local network alive, deploy the smart contract there and then interact with it. You can think on this like a local web server. When you develop an API you have a local server running it so the client application can talk to it so you can develop and test, this is the same idea.

Head over to your terminal and run tthe following command.

```bash
$ npm run hardhat node
```

This will ramp up a local ethreum network using hardhat and will show in the console a set of accounts and private keys that can be used for testing pruposes.

These are 20 test accounts and addresses created for your. Each account is also rich! are loaded with 10000 ETH. (fake eth btw).

Right now this is just an empty blockchain. There are no blocks nor smart contracts on it. But you want to create a new block and put the smart contract that you just created there, for that you need to create a little script that will use ethers.js power to compile and deploy.

Deploy a Solidity Smart Contract with Hardhat to a Local Ethereum Network

To reduce the amount of gas used, we will create a withdraw function that lets the owner of the contract withdraw funds that have been added when they want rather than withdraw with every transaction.

Your smart contract is ready to be used, since you have confidence in the code after testing it.

But there are a few things that you can add to improve the behavior and security of it.

Let's start by adding events.

Similar to any other programming language Events in solidity are a way to inform the calling application about the current state of the contract. Like notifying the javascript client about a change made to the contract allowing the client application to react to it.

Events are defined within the contract as a global and are called from inside a function. To declare an event you use the `event` keyword, followed by an identifier and the parameter list. The values passed to the parameters are then used to log the information that is saved as part of the transaction inside the block.

To send the event you'll use the `emit` keyword from anywhere inside the contract by passing the name of the event and the corresponding parameters.

For this contract you want to let the client application to know when a tip was successfully sent. To do that you'll create a new event named `NewTip` like this

```javascript
event NewTip(address indexed from, string message, string name, uint256 amount);
```

It will accept an address to identify who sent the tip, the message of the tip and also the name and the amount of eth sent,

And then, it will be emitted from the `sendTip` function

```javascript
// send the event
emit NewTip(msg.sender, _message, _name, msg.value);
```

This event will use the same parameters that the rest of the `sendTip` operation.

- `msg.sender` to get the address of who sent the tip
- `_message` the string with the message sent
- `_name` the name of the sender
- `msg.value` the amount of `eth` sent.

Add eth Withdraw Functionality in a Smart Contract and Emit Solidity Events

Tests are important, as always but it may be more important when you are developing an smart contract. Why?
Because the blockchain is immutable.

The main difference between Solidity development and other languages or environment is that after the implementation of an smart contract in the Ethereum network the contract is immutable, meaning that it will never be modified or updated again.

The first code that get's deployed is the one that will remains unchanged for the rest of time. This is because one the main concerns of Solidity is security. If there's a flaw in your contract code, there is no way to patch it later. You would have to tell to the users of the contract to use a different contract (identified by an `address`) to get the fix.

Sounds uncomfortable right? But, believe it or not, this is a **feature**. The code that your wrote it becomes a **law**. IF you read the code of an smart contract (and verified that that code is actually the one deployed) you can be sure that every time you call a function it will behave exactly as the code says it.

So you need to be sure before deploy and the best way to get that confidence is by testing.

[Hardhat](https://hardhat.org/) (the development environment of choice for this course) offers the tools to test the smart contract using Javascript or Typescript by leveraging the power of [Ethers.js](https://docs.ethers.io/) to interact with the Ethereum network and [Waffle](https://getwaffle.io/), a testing library built on top of Ethere.js

You already dip your toes on a bit of testing code, now let's add a more robust test to check that the contract behaves as expected.



Test Smart Contract Functionality with Ethers and Hardhat

It's time to add functionality to our smart contract. In this lesson we will define a function `sendTip` that accepts a `_message` and `_name`. We will check the senders balance and if it's higher than the tip amount transer the eth specified. We will also increment our `totalTips` contract variable and  add the tip to our Tips array.

While adding this function you'll learn how to handle errors within Solidity functions

Create a Smart Contract Function that Transfers eth from One Address to Another

Now, we have the base contract to work on. The goal of the contract is to act as a tip jar, a way to allow users or followers to send you some ETH through the web, including some message. But also, to showcase the people that sent you that money by making the actions public in the web too. This means that the contract need to:

1. Identify to who to send the tip
2. Store the tip sender data including the message, and the amount of eth.
3. Allow a way to retrieve the actions (the tips sent) made through it.

We will add an `address payable` variable of owner to tell the contract that the contract will receive payments. 

For the actual tip data, we will define a Tip `struct` that will have all the information needed to send a tip.

Store Data in a Smart Contract struct Variable

The first step in this journey will be get your hands into a new language to harness the power of the blockchain. I'm talking about Solidity

**This isn't a Solidity focused course but we will review some concepts to get you up to speed with it.

The first thing to notice is that Solidity is an statically-typed language created specifically to create smart contracts for the Ethereum network.

An smart contract is like "real life" contract. Is a way to establish terms for an agreement. But the "smart" part it's because this type of contract is declared as machine code that is executed in a blockchain. This expand the idea of the original blockchain, Bitcon, aout sending and receiveing money without a centralized intermiediary to create trust and make possible to automate and dcentralize virtually any kinf of deal or transaction.

Since the contracts run in the blockchain (like Ethereum network) they offer high security, trust, inmmutability and reliability.

Since the blockchain is immutable you have to be care on what are you deploying. After deployed it cannot be change and you'll need to deploy a new contract.

For Solidity, a contract is a collection of code (a program) and data (the state) that lives inside the blockchain, it can be identify by its Ethereum address. You can leverage your previous knowledge and use some analogies to understand the smart contract.

You can think on the contract as your "backend" or "server side" program. This server expose an API and store some data (the state) into the a databse (the blockchain), but the database is not written nor query with a different language like SQL, you can imagine that Solidity is automatically connected to the database and everything you deeclare (like a variable) is stored in the database.

Create and Test a Solidity Smart Contract with a Public Variable

We'll start our project by installing all of the necessary tools to get started with Web3 development. 

This project will use SvelteKit Tailwind CSS for it's UI. We have some handy initializers that we will run to do the configuration for us. 

On the Web3 side of things we will use Hardhat, Ethers.js, and hardhat-waffle for smart contract development.

Set up an Ethereum Project using SvelteKit with Tailwind CSS and Hardhat

Hardcoding your script to search for only one type of file (like `.md`) is a major limitation. This lesson shows you how to make your script smarter by teaching it to infer the desired file type directly from the user's natural language prompt.

You'll implement a two-stage AI process. The first stage analyzes the user's request to determine the file extension they're interested in (e.g., `md`, `txt`). The second stage then uses that file type to dynamically search the file system, find the relevant files, and execute the command.

**Workflow demonstrated in this lesson:**

* Add a preliminary `generateObject` call to extract a `fileType` from the user's initial prompt.
* Define a list of supported `fileTypes` (e.g., `['md', 'txt']`) and use a detailed prompt to guide the AI, instructing it to choose from the list or use a default.
* Dynamically construct the `globby` pattern using the inferred `fileType` (e.g., `*.$[fileType]`).
* Implement error handling to inform the user if they request an unsupported file type or if no files of the specified type are found.
* Demonstrate how the script defaults to searching for Markdown files but can correctly search for other types when specified (e.g., "summarize my profile from my pdf").

This technique adds a significant layer of intelligence to your script, allowing it to adapt its behavior based on the user's intent and making it a far more flexible and intuitive command-line tool.

John Lindquist

Dynamically Inferring File Types from User Prompts in Local AI Scripts

Hardcoding command logic and prompts directly into your script makes it rigid and difficult to extend. This lesson demonstrates a more flexible and powerful pattern: defining your script's commands as individual Markdown files in a dedicated directory.

This approach allows you to easily add, remove, or modify commands without touching the core script logic. You'll learn how to dynamically load these command files, present them as options to the AI, and inject the corresponding instructions into your prompt, creating a truly modular and extensible tool.

**Workflow demonstrated in this lesson:**

* Create a `commands` directory to house your command definitions (e.g., `summarize.md`, `translate.md`).
* In each file, write the specific prompt or instructions for that command.
* Use `globby` to discover all available command files at runtime.
* Create a `commandMap` that maps the filename (e.g., "summarize") to its file content.
* Update the Zod schema to use an `enum` of the dynamically discovered command names.
* Dynamically construct the final text generation prompt by injecting the instructions from the user-selected command file.
* Add a `profile.md` file to demonstrate how the script can act on user-created content with the new file-based commands.

By the end of this lesson, your script will be architected for easy expansion, allowing you to build a robust library of custom commands simply by adding new text files.

Creating Dynamic, File-Based Commands for Local AI Scripts

AI models won't always give you the answer you expect; sometimes they'll return invalid data or "hallucinate" a response that doesn't fit your schema. This lesson teaches you how to make your local AI script more robust by implementing essential error handling for user commands.

You'll learn how to validate the AI's output against a predefined list of valid commands and gracefully catch errors when the model generates an unexpected response, ensuring your script fails predictably and provides helpful feedback to the user.

**Workflow demonstrated in this lesson:**

* Define a list of `validCommands` (e.g., `summarize`, `translate`, `review`) to constrain the AI's choices.
* Wrap the `generateObject` call in a `try...catch` block to handle cases where the AI's output doesn't match the Zod schema.
* Implement a type guard function (`isValidCommand`) to check if the command returned by the AI is in your list of valid commands.
* Provide clear error messages to the user, indicating what went wrong and what the valid commands are.
* Test the script with invalid commands (like "taco" or "delete") to confirm that the error handling works as expected.

By adding these checks, you make your script significantly more reliable and user-friendly, preventing it from crashing on bad input and guiding the user toward correct usage.

Handling Invalid Commands and AI Errors in Local Scripts

Once your script can understand a user's command and identify a target file, the next step is to perform an action. This lesson shows you how to use the Vercel AI SDK's `generateText` function to process a file's content and produce a relevant, AI-generated response.

You'll learn how to read the content from the file path you've already identified, combine it with the inferred command into a new, more detailed prompt, and then pass it back to your local AI model to generate a final text output, such as a summary or a review.

**Workflow demonstrated in this lesson:**

* Import the `generateText` function from the `ai` SDK.
* Read the file content from the `filePath` returned by `generateObject`.
* Construct a new, structured prompt that includes the original command (e.g., "summarize") and the full content of the file.
* Pass this new prompt to the `generateText` function.
* Log the resulting AI-generated text to the console.
* Run the script with commands like "Summarize our TypeScript config" or "Review my package file" to see the full, context-aware output.

This lesson completes the core cycle of your AI script: understanding a command, reading the relevant context, and generating a useful, text-based result, all on your local machine.

Generating Text with a Local AI Model Based on File Content

A command is useless without knowing what to act upon. This lesson elevates your local AI script by giving it context from your project's file system. You'll learn how to find all relevant files and empower the AI to intelligently select the correct one based on your prompt.

The core technique involves using the `globby` package to scan for files and then feeding that list into a Zod `enum`. This constrains the AI's choices to a valid set of files, allowing it to accurately determine which file you're referring to in a natural language command.

**Workflow demonstrated in this lesson:**

* Install and import the `globby` library to find files based on a pattern (e.g., all `.json` files).
* Add a `filePath` field to your Zod schema.
* Define the `filePath` as `z.enum(files)`, where `files` is the array of file paths returned by `globby`.
* Pass a natural language prompt like "Summarize our typescript config" to the script.
* Observe how the AI correctly identifies `tsconfig.json` from the list of available JSON files and returns it as structured data.

This lesson is a crucial step toward building powerful, context-aware tools that can intelligently interact with your project's files based on simple, human-readable commands.

Providing File System Context to Local AI Scripts with Globby and Zod Enums

Hardcoded prompts are great for testing, but a truly useful script needs to respond to dynamic user input. This lesson shows you how to make your local AI script interactive by capturing command-line arguments.

You'll learn how to use Node.js's built-in `parseArgs` utility to take a user's natural language command directly from the terminal and pass it to your local language model for processing.

**Workflow demonstrated in this lesson:**

* Import the `parseArgs` function from the `node:util` module.
* Configure `parseArgs` to accept positional arguments from the command line.
* Replace the static, hardcoded prompt with the dynamic input provided by the user.
* Run the script with different sentences (e.g., `"Download a file"` or `"I want to download a file"`) to demonstrate that the AI can correctly infer the core command regardless of the phrasing.

By the end of this short lesson, your script will be able to accept and process commands directly from the user, making it a flexible and reusable command-line tool.

Processing Dynamic User Input in Local AI Scripts

While command-line chats with local AI models are useful, the real power comes from scripting. This lesson walks you through the initial setup for creating powerful, local AI-driven scripts by combining Ollama, the Vercel AI SDK, and Zod.

You'll learn how to take a natural language command, like "Concatenate all of the markdown files in my Downloads folder into a single file," and use the `generateObject` function to intelligently extract structured, actionable data from it.

**Workflow demonstrated in this lesson:**

* Install and set up the `ollama-ai-provider`, `ai`, and `zod` packages.
* Create a local AI model instance using Ollama and the Gemma 3 model.
* Define a Zod schema to specify the exact data structure you want to extract (e.g., `{ command: string }`).
* Use `generateObject` to pass a natural language prompt to the local model and receive a structured JSON object that matches your schema.
* Run the script using `bun` to see the local model successfully infer the "concatenate" command from a full sentence.

This foundational lesson unlocks the ability to build sophisticated, automated workflows by translating human language into machine-readable commands, all running entirely on your local machine.

Scripting Local Language Models with Ollama and the Vercel AI SDK

### **Build Powerful, Local AI-Driven Command-Line Scripts**

Go beyond simple AI chats and unlock the true potential of local language models by learning to script them. This course will guide you step-by-step through the process of building a flexible and intelligent command-line tool that can understand natural language, interact with your file system, and automate complex tasks—all running entirely on your local machine.

Starting with a blank TypeScript file, you'll learn how to combine the power of **Ollama**, the **Vercel AI SDK**, and **Zod** to create a script that's not just functional, but smart. We'll begin by teaching the AI to parse a simple command from a sentence. From there, we'll progressively add layers of sophistication: processing dynamic user input, making the script aware of your project's files, and generating meaningful text-based output.

You'll also learn critical patterns for building robust and extensible tools. We'll cover how to handle invalid commands and unexpected AI responses gracefully. Then, you'll refactor the script to use a modular, file-based system for its commands, allowing you to easily add new functionality without touching the core logic. Finally, we'll add a layer of intelligence that allows your script to infer exactly what you mean, right down to the type of file you want it to process.

By the end of this course, you will have built a powerful, personalized AI assistant and gained the foundational skills to create sophisticated local automations for any task you can imagine.

**What You'll Learn:**

* How to script local language models using Ollama and the Vercel AI SDK.
* To parse and process dynamic user input from the command line.
* To use Zod schemas, including `enums`, to validate and structure AI model output.
* Techniques for providing file-system context to your scripts with `globby`.
* How to generate context-aware text based on file content.
* Robust error handling for invalid user commands and unexpected AI responses.
* To build a modular and extensible tool by defining commands in separate files.
* To intelligently infer user intent, such as dynamically determining file types from a prompt.

Build a local AI-powered command-line assistant from scratch. Learn to use Ollama, Vercel AI SDK & Zod to parse input, automate tasks, and run scripts offline.

This lesson demonstrates how to give Cursor direct access to your terminal output using the [Nx MCP extension](https://nx.dev/getting-started/ai-integration?utm_source=youtube&utm_medium=videos&utm_campaign=Nx-AI&utm_id=Nx-AI), enabling automatic error detection and fixing in monorepo projects.

**Workflow demonstrated in this lesson:**

- Encounter a runtime error when serving a Vite application in an Nx monorepo
- Install and configure the Nx Console extension with MCP integration
- Ask Cursor to check your terminal for errors and fix the underlying issues
- Watch as the AI accesses running task details and terminal output through Nx MCP tools
- Observe how it analyzes workspace relationships to identify incorrect imports and missing dependencies

**Key benefits:**

- Seamless Error Handling: AI gains direct access to terminal output and running task information
- Automatic Diagnosis: Accurately identify the root cause of errors without manual copying
- Context-Aware Fixes: Leverage workspace knowledge to implement proper solutions
- Reduced Context Switching: Eliminate the need to manually relay information between terminal and AI



Juri Strumpflohner

Enable Cursor to directly access the terminal to read the output and fix issues

See how Nx MCP gives AI coding assistants direct access to terminal output and running tasks, enabling automatic error detection and fixing.

This lesson demonstrates how to solve the predictability problem when using AI to generate code in monorepos by combining Cursor with Nx generators.

**Workflow demonstrated in this lesson:**

- Explore how Nx generators provide predictable scaffolding for new libraries
- Set up [Nx Console with MCP](https://nx.dev/features/enhance-AI?utm_source=egghead&utm_medium=videos&utm_campaign=Nx-AI&utm_id=Nx-AI) integration to expose generator capabilities to Cursor
- Use the AI assistant to brainstorm a new feature with the help of the Nx MCP tools
- Have Cursor leverage the Nx MCP tools to trigger the appropriate Nx generator for creating a React-based feature library
- Observe how the AI assistant takes the generated library and further integrates it into the codebase

**Key benefits:**

- **Predictable Code Structure**: Ensure generated libraries follow your project's and organization's conventions
- **Consistent Setup**: Maintain uniform configuration across TypeScript projects and package.json files
- **Guided Implementation**: Combine AI instructions with structured, predictable code generation

Enable Cursor to generate predictable project setups

Learn how to combine Cursor AI with Nx generators for predictable library creation in monorepos, ensuring consistent structure across your codebase.

Search

Search Results