Goyquest

Lightweight CLI tool that reacts to Discord channel messages with an emoji (🇮🇱 by default) using the raw Discord HTTP API and Gateway WebSocket — no discord.py.

What it does

• Live (default) — watches for new messages via WebSocket and reacts in real time
• Backfill (optional) — react to recent messages by time (--minutes) or message count (--last)
• Run timer (optional) — stop automatically after N minutes, or run until Ctrl+C (default)
• Flexible targeting — one channel, comma-separated channels, or --channel all for every accessible text channel in a server
• Permission-aware — skips channels where your account cannot add reactions (no repeated 403 spam)

Requirements

• Python 3.10+
• Dependencies in requirements.txt
• Your Discord user token and server ID
• A channel ID, comma-separated channel IDs, or all for --channel

Setup

cd goyquest
python -m pip install -r requirements.txt

Getting your token

To get the token for your personal account:

Automating user accounts is technically against TOS — use at your own risk!

1. Open Discord in your web browser and login
2. Open any server or direct message channel
3. Press Ctrl+Shift+I to show developer tools
4. Navigate to the Network tab
5. Press Ctrl+R to reload
6. Switch between random channels to trigger network requests
7. Search for a request that starts with messages
8. Select the Headers tab on the right
9. Scroll down to the Request Headers section
10. Copy the value of the authorization header

Getting channel and server IDs

You always need a server ID for --server. You only need channel IDs if you are not using --channel all.

1. Open Discord (desktop app or web browser)
2. Click the gear icon next to your username to open User Settings
3. Go to Advanced in the left sidebar
4. Turn on Developer Mode
5. Close settings

Server ID:

1. In the server list on the left, right-click the server icon (the circular image)
2. Click Copy Server ID
3. Paste it into --server (e.g. --server 1111111111111111111)

Channel ID (skip if using --channel all):

1. Right-click the text channel name in the channel list
2. Click Copy Channel ID
3. Paste it into --channel (e.g. --channel 1234567890123456789)

For multiple channels, copy each channel ID and separate them with commas:

--server 1111111111111111111 --channel 1234567890123456789,9876543210987654321

Usage

Single channel (live only — default):

python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789

Or:

.\run.ps1 --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789

Multiple channels (comma-separated):

python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789,9876543210987654321

All text channels in the server:

python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel all

--channel all (alias: --channels all) discovers every text and announcement channel in the server. Channels you cannot read are skipped during setup. Channels you can read but cannot react in are skipped after the first permission error.

Run timer (optional):

python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789                # run until Ctrl+C (default)
python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789 --timer 30   # stop after 30 minutes

The timer counts from script start and applies to backfill and live listening.

Backfill by time (optional, in minutes):

python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789                    # live only (default)
python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789 --minutes 60   # backfill last 60 minutes, then live
python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789 --minutes 1440 # backfill last 24 hours, then live

Backfill by message count:

python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789 --last 30
python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel all --last 20

Use --minutes or --last, not both. Both apply per channel. Use --delay to slow backfill if you hit rate limits (e.g. --delay 1).

Backfill only (no WebSocket):

python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789 --last 30 --backfill-only

Backfill, then live, with a run timer:

python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789 --minutes 60 --timer 120

Backfills the last 60 minutes, then listens for new messages, and stops after 120 minutes total (including backfill).

Validate token and server access (no reactions, no WebSocket):

python react_http.py --token YOUR_TOKEN --server 1111111111111111111 --channel 1234567890123456789 --backfill-only

Checks your token, server, and channel access via the REST API, then exits. With --channel all, this resolves all accessible channels without backfilling (unless --minutes or --last is also set).

CLI flags

Flag	Default	Description
--token	(required)	User token
--channel, --channels	(required)	Channel ID(s), comma-separated, or all
--guild, --server	(required)	Server ID — channels must belong to this server
--emoji	🇮🇱	Emoji to react with
--minutes	0	Backfill messages from the last N minutes (0 = skip)
--last	0	Backfill the last N messages per channel (0 = skip)
--delay	0.35	Seconds between reactions during backfill
--skip-bots	off	Ignore messages from bots
--skip-self	off	Skip your own messages
--backfill-only	off	Backfill then exit (no live WebSocket)
--timer	0	Stop after N minutes (0 = run until Ctrl+C)

--minutes and --last are mutually exclusive.

Environment variables

Environment variable	CLI flag
DISCORD_TOKEN	--token
DISCORD_CHANNEL	--channel (comma-separated IDs or all)
DISCORD_GUILD	--server / --guild
DISCORD_EMOJI	--emoji
BACKFILL_MINUTES	--minutes
BACKFILL_LAST	--last
RUN_MINUTES	--timer

If DISCORD_TOKEN, DISCORD_GUILD, and DISCORD_CHANNEL are set, you can omit --token, --server, and --channel on the command line. Put secrets in a .env file locally — it is gitignored.

How it works

Phase	Method
Auth	Authorization: <token> on REST requests (Chrome user-agent)
Backfill	GET /channels/{id}/messages → PUT .../reactions/{emoji}/@me
Live	Gateway WebSocket MESSAGE_CREATE → PUT reaction
Rate limits	On HTTP 429, waits for Discord's retry_after before retrying
Permissions	On HTTP 403 missing react permission, skips that channel for the rest of the run

Project layout

goyquest/
├── react_http.py    # Main script
├── requirements.txt
├── run.ps1          # PowerShell launcher
├── .gitignore
└── README.md

Random findings

Anecdotal things we've run into while using this — not official Discord policy, but worth knowing:

Join servers before you run, not one-by-one after kicks

If you get kicked or removed from a channel (or a couple of them), Discord may temporarily restrict your account from joining new servers or channels for roughly 24–48 hours. That lockout is easy to hit if you join a server, run the script, get kicked, then immediately join another server and repeat.

Safer approach:

1. Join all the servers you plan to use first
2. Let your account settle in those servers
3. Then run Goyquest across the channels you already have access to

Avoid the loop of "join server → run script → get kicked → join next server" in quick succession.

Notes

• Automating a user account may violate Discord's Terms of Service. Use at your own risk.
• Do not commit tokens to public repositories.
• Press Ctrl+C in the terminal to stop early (even when --timer is set).
• Reading messages and adding reactions require different permissions. A channel may appear in --channel all but still be skipped if your role cannot add reactions.
• Discord rate limits are dynamic. If you see frequent slowdowns or 429 responses during backfill, increase --delay (try 1–2 seconds).
• Windows PowerShell may not display 🇮🇱 in the console; reactions still work on Discord.