added readme

README.md

@@ -1 +1,84 @@
 ![og image](https://supermemory.dhr.wtf/og-image.png)
+
+# SuperMemory
+
+## 👀 What is this?
+
+Build your own second brain with supermemory. It's a ChatGPT for your bookmarks. Import tweets or save websites and content using the [chrome extension](https://chromewebstore.google.com/detail/supermemory/afpgkkipfdpeaflnpoaffkcankadgjfc?hl=en-GB&authuser=0) (the extension on webstore is not updated, please use the one in the repo)
+
+Well, here's the thing - me and @yxshv save a _lot_ of content on the internet.
+
+Twitter bookmarks, websites, snippets, etc.
+
+But we never look back to it - to us, it's like throwing information in the void.
+
+Supermemory fixes this.
+
+## How do I use this?
+
+Just go to [supermemory.dhr.wtf](https://supermemory.dhr.wtf) and sign in with your google account.
+
+To use the chrome extension, please download it from the [releases page](https://github.com/dhravya/supermemory/releases) and load it in chrome.
+
+## 👨‍💻 The Stack
+
+![overview](https://i.dhr.wtf/r/Clipboard_Apr_14,_2024_at_4.52 PM.png)
+
+Supermemory has three main modules, managed by [turborepo](https://turbo.build):
+
+#### `apps/web`: The main web UI.
+
+The database, auth etc logic is here
+
+![App preview](https://i.dhr.wtf/r/Clipboard_Apr_14,_2024_at_4.10 PM.png)
+
+Built with:
+
+- Nextjs 14
+- [Next Auth](https://next-auth.js.org/)
+- [Drizzle ORM](https://drizzle.team/)
+- [Cloudflare D1 database](https://developers.cloudflare.com/d1/get-started/)
+- Cloudflare ratelimiter
+- [TailwindCSS](https://tailwindcss.com)
+- [shadcn-ui](https://ui.shadcn.com)
+- And some other amazing open source projects like [Novel](https://novel.sh) and [vaul](https://vaul.emilkowal.ski/)
+- Hosted on Cloudflare Pages
+
+#### `apps/extension`: Chrome extension
+
+The chrome extension is one of the most important part of the setup, but is not required.This is to easily add pages to your memory.
+
+![Chrome extension preview](https://i.dhr.wtf/r/Clipboard_Apr_14,_2024_at_3.54 PM.png)
+
+You can also use it to import all your twitter bookmarks!
+![Import bookmarks](https://i.dhr.wtf/r/Clipboard_Apr_14,_2024_at_3.56 PM.png)
+
+Built with:
+
+- [CRXJS](https://crxjs.dev/vite-plugin/getting-started/react/create-project)
+- Vite
+- [TailwindCSS](https://tailwindcss.com)
+- [shadcn-ui](https://ui.shadcn.com)
+- React
+
+#### `apps/cf-ai-backend`: This module handles the vector store and AI response generation
+
+This is where the magic happens!
+Built with:
+
+- Cloudflare Workers
+- [Cloudflare AI](https://ai.cloudflare.com)
+- [Cloudflare Vectorize](https://developers.cloudflare.com/vectorize/)
+- [Cloudflare Queues](https://developers.cloudflare.com/queues/)
+- [Cloudflare Browser Rendering](https://developers.cloudflare.com/browser-rendering/)
+- [Cloudflare KV](https://developers.cloudflare.com/kv)
+
+## Contribute or self host
+
+Supermemory is design to be set up easily locally and super duper easy to set up 💫
+
+Please see the [SETUP-GUIDE.md](SETUP-GUIDE.md) for setup instructions.
+
+### Contributing
+
+Contributions are very welcome! A contribution can be as small as a ⭐ or even finding and creating issues.

SETUP-GUIDE.md

@@ -0,0 +1,77 @@
+# Setup guide
+
+## Prerequisites
+
+- [bun](https://bun.sh/)
+- [wrangler](https://developers.cloudflare.com/workers/cli-wrangler/install-update)
+
+## Steps
+
+1. Clone the repo
+2. Run `bun install` in the root directory
+3. Create a `.dev.vars` file in `apps/web` with the following content:
+
+```bash
+GOOGLE_CLIENT_ID="-"
+GOOGLE_CLIENT_SECRET="-"
+NEXTAUTH_SECRET='nextauthsecret'
+DATABASE_URL='database.sqlite'
+NEXTAUTH_URL='http://localhost:3000'
+BACKEND_SECURITY_KEY='veryrandomsecuritykey'
+```
+
+4. Setup the database:
+
+First, edit the `wrangler.toml` file in `apps/web` to point the d1 database to your account.
+
+You can create a d1 database by running this command
+
+```
+wrangler d1 create DATABASE_NAME
+```
+
+And then replace these values
+
+```
+[[d1_databases]]
+binding = "DATABASE"
+database_name = "YOUR_DATABASE_NAME"
+database_id = "YOUR_DB_ID"
+```
+
+Simply run this command in `apps/web`
+
+```
+wrangler d1 execute dev-d1-anycontext --local --file=db/prepare.sql
+```
+
+If it runs, you can set up the cloud database as well by removing the `--local` flag.
+
+5. You need to host your own worker for the `apps/cf-ai-backend` module.
+
+To do this, first edit the `.dev.vars` file in `apps/cf-ai-backend` with the following content:
+
+```bash
+SECURITY_KEY ="veryrandomsecuritykey"
+// Why? to generate embeddings with 4000+ tokens
+OPENAI_API_KEY="sk-"
+```
+
+6. Run this command to initialise vector database
+
+```
+wrangler vectorize create --dimensions=1536 supermem-vector-1 --metric=cosine
+```
+
+7. Change the `wrangler.toml` file in `apps/cf-ai-backend` to point to your KV namespace
+
+8. Run `bun dev` in the root directory and Voila! You have your own supermemory instance running!
+
+> Note: You need to replace the url `https://cf-ai-backend.dhr.wtf` everywhere with your own url for the cf-ai-backend module.
+
+## Deploying
+
+To deploy the web app, run `bun deploy` in the `apps/web` directory.
+To deploy the cf-ai-backend module, run `wrangler publish` in the `apps/cf-ai-backend` directory.
+
+To get the extension running, you need to build it first. Run `bun build` in the `apps/extension` directory and then load the extension in chrome.

apps/cf-ai-backend/wrangler.toml

@@ -27,51 +27,3 @@ binding = "MYBROWSER"
 
 [placement]
 mode = "smart"
-
-# Variable bindings. These are arbitrary, plaintext strings (similar to environment variables)
-# Note: Use secrets to store sensitive data.
-# Docs: https://developers.cloudflare.com/workers/platform/environment-variables
-# [vars]
-# MY_VARIABLE = "production_value"
-
-# Bind a KV Namespace. Use KV as persistent storage for small key-value pairs.
-# Docs: https://developers.cloudflare.com/workers/runtime-apis/kv
-# [[kv_namespaces]]
-# binding = "MY_KV_NAMESPACE"
-# id = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
-
-# Bind an R2 Bucket. Use R2 to store arbitrarily large blobs of data, such as files.
-# Docs: https://developers.cloudflare.com/r2/api/workers/workers-api-usage/
-# [[r2_buckets]]
-# binding = "MY_BUCKET"
-# bucket_name = "my-bucket"
-
-# Bind a Queue producer. Use this binding to schedule an arbitrary task that may be processed later by a Queue consumer.
-# Docs: https://developers.cloudflare.com/queues/get-started
-# [[queues.producers]]
-# binding = "MY_QUEUE"
-# queue = "my-queue"
-
-# Bind a Queue consumer. Queue Consumers can retrieve tasks scheduled by Producers to act on them.
-# Docs: https://developers.cloudflare.com/queues/get-started
-# [[queues.consumers]]
-# queue = "my-queue"
-
-# Bind another Worker service. Use this binding to call another Worker without network overhead.
-# Docs: https://developers.cloudflare.com/workers/platform/services
-# [[services]]
-# binding = "MY_SERVICE"
-# service = "my-service"
-
-# Bind a Durable Object. Durable objects are a scale-to-zero compute primitive based on the actor model.
-# Durable Objects can live for as long as needed. Use these when you need a long-running "server", such as in realtime apps.
-# Docs: https://developers.cloudflare.com/workers/runtime-apis/durable-objects
-# [[durable_objects.bindings]]
-# name = "MY_DURABLE_OBJECT"
-# class_name = "MyDurableObject"
-
-# Durable Object migrations.
-# Docs: https://developers.cloudflare.com/workers/learning/using-durable-objects#configure-durable-object-classes-with-migrations
-# [[migrations]]
-# tag = "v1"
-# new_classes = ["MyDurableObject"]

apps/web/wrangler.toml

@@ -2,10 +2,6 @@ name = "web"
 compatibility_date = "2024-03-29"
 compatibility_flags = ["nodejs_compat"]
 
-[[vectorize]]
-binding = "VECTORIZE_INDEX" 
-index_name = "anycontext-idx"
-
 [[d1_databases]]
 binding = "DATABASE"
 database_name = "dev-d1-anycontext"