Read something but forget where it was? Seen remembers what you have seen.

Seen is a simple personal knowledge management tool made for impatient people like me.

Features

• Dead simple, just send a link, and search later for whatever you still remember.

• Just a simple telegram bot, no bloatware.

• Cures your allergic to AI by using a lot of AI in this project: document processing, vector search, etc.

Demo

Roadmap

Seen is migrating to use VectorLite as its vector database, allowing lower latency and more reliable service.

As the natural next step, we will offer a hosted version of Seen.

Hosted service

Let me know if you are interested in a hosted version of Seen.

Self-hosting

Setting up the self-hosted version is a bit complicated. Please let me know if there are any issues. Self-hosting uses Cloudflare free-tier, should be more than enough for personal use.

1. Create a Telegram Bot

Create a Telegram bot (from @BotFather) and get an API token.

2. Configure Cloudflare Workers

Login to Cloudflare:

npx wrangler login

Clone this repository

git clone https://github.com/XiangpengHao/seen.git
cd seen

Setup Telegram

1. Add your Telegram bot token to Cloudflare Workers as a secret:

   npx wrangler secret put BOT_TOKEN

   When prompted, paste your Telegram bot token.

2. Get your Chat ID by sending a message to your bot and then visiting:

   https://api.telegram.org/bot<BOT_TOKEN>/getUpdates
   

   Replace <BOT_TOKEN> with your actual bot token. Look for the chat object in the response and find your id.

3. Add your Chat ID to the environment variables in your wrangler.toml file:

   [vars]
   AUTHORIZED_CHAT_IDS = "YOUR_CHAT_ID"

   Replace YOUR_CHAT_ID with the chat ID you obtained in the previous step. For multiple authorized users, separate IDs with commas: "ID1,ID2,ID3".

Setup CloudFlare

Add your CloudFlare account ID and API token to Cloudflare Workers as secrets:

npx wrangler secret put CF_ACCOUNT_ID
npx wrangler secret put CF_API_TOKEN

Setup Gemini

Seen use Gemini to process documents. You can get a free API key from Gemini API.

Add your Gemini API key to Cloudflare Workers as a secret:

npx wrangler secret put GEMINI_API_KEY

Setup D1 database

npx wrangler d1 create seen

Now you should copy the output to the corresponding field in wrangler.toml, e.g.

[[d1_databases]]
binding = "SEEN_DB"
database_name = "seen"
database_id = "bba483bb-4377-46f1-a8d4-df83772edc95"

Configure the database:

npx wrangler d1 execute seen --command "SQL_BELOW" --remote

CREATE TABLE IF NOT EXISTS links (
    id TEXT PRIMARY KEY,
    url TEXT NOT NULL,
    created_at TEXT NOT NULL,
    bucket_path TEXT NOT NULL,
    content_type TEXT NOT NULL,
    size INTEGER NOT NULL,
    title TEXT NOT NULL,
    summary TEXT NOT NULL,
    chunk_count INTEGER NOT NULL
);
CREATE INDEX IF NOT EXISTS idx_links_url ON links(url);
CREATE INDEX IF NOT EXISTS idx_links_id ON links(id);

Setup Vectorize

npx wrangler vectorize create seen-index --dimensions=768 --metric=cosine

3. Build and Deploy to Cloudflare Workers

Build and deploy your bot to Cloudflare Workers:

env RUSTFLAGS='--cfg getrandom_backend="wasm_js"' npx wrangler deploy

This will deploy your bot and give you a URL (something like https://your-bot.your-username.workers.dev).

4. Set up Telegram Webhook

Tell Telegram where to send updates by setting up a webhook. Open a browser and navigate to:

https://api.telegram.org/bot<BOT_TOKEN>/setWebhook?url=https://your-bot.your-username.workers.dev/webhook

Replace <BOT_TOKEN> with your actual bot token and update the Worker URL accordingly.

You should see a response like:

{"ok":true,"result":true,"description":"Webhook was set"}

License

MIT