DiscordSAFE
MCP server for discord bot
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://smithery.ai/server/@hanweg/mcp-discord) A Model Context Protocol (MCP) server that provides Discord integration capabilities to MCP clients like Claude Desktop.
Available Tools
Server Information
list_servers: List available serversget_server_info: Get detailed server informationget_channels: List channels in a serverlist_members: List server members and their rolesget_user_info: Get detailed information about a user
Message Management
send_message: Send a message to a channelread_messages: Read recent message historyadd_reaction: Add a reaction to a messageadd_multiple_reactions: Add multiple reactions to a messageremove_reaction: Remove a reaction from a messagemoderate_message: Delete messages and timeout users
Channel Management
create_text_channel: Create a new text channeldelete_channel: Delete an existing channel
Role Management
add_role: Add a role to a userremove_role: Remove a role from a user
Installation
- Set up your Discord bot:
- Create a new application at Discord Developer Portal
- Create a bot and copy the token
- Enable required privileged intents:
- MESSAGE CONTENT INTENT
- PRESENCE INTENT
- SERVER MEMBERS INTENT
- Invite the bot to your server using OAuth2 URL Generator
- Clone and install the package:
# Clone the repository git clone https://github.com/hanweg/mcp-discord.git cd mcp-discord # Create and activate virtual environment uv venv .venv\Scripts\activate # On macOS/Linux, use: source .venv/bin/activate ### If using Python 3.13+ - install audioop library: `uv pip install audioop-lts` # Install
ced9701e70afOBSERVED · 2026-10-06Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add mcp-discord --env DISCORD_TOKEN=${DISCORD_TOKEN} -- uvx mcp-discord{
"mcpServers": {
"mcp-discord": {
"command": "uvx",
"args": [
"mcp-discord"
],
"env": {
"DISCORD_TOKEN": "${DISCORD_TOKEN}"
}
}
}
}Exposed tools (15)
6 read · 5 write · 4 destructive. Blast radius: 4 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
add_multiple_reactions | write | Add multiple reactions to a message |
add_reaction | write | Add a reaction to a message |
add_role | write | Add a role to a user |
create_text_channel | write | Create a new text channel |
delete_channel | destructive | Delete a channel |
get_channels | read | Get a list of all channels in a Discord server |
get_server_info | read | Get information about a Discord server |
get_user_info | read | Get information about a Discord user |
list_members | read | Get a list of members in a server |
list_servers | read | Get a list of all Discord servers the bot has access to with their details such as name, id, member count, and creation date. |
moderate_message | destructive | Delete a message and optionally timeout the user |
read_messages | read | Read recent messages from a channel |
remove_reaction | destructive | Remove a reaction from a message |
remove_role | destructive | Remove a role from a user |
send_message | write | Send a message to a specific channel |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- declared (1 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (2)
delete_channel, moderate_message, remove_reaction, remove_role
discord.py, mcp
Gates applied: no_behavioural_pass.
ced9701e70affull audit observations/trust-audit/mcp-server/hanweg__discord-1.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-06 | ced9701e70af | SAFE | B | 89 | first audit |
Questions
What is the Discord MCP server?
MCP server for discord bot
What tools does Discord expose?
15 in total: 6 read-only, 5 that write, and 4 that can delete or overwrite (delete_channel, moderate_message, remove_reaction, remove_role). Every one is listed on this page with its risk.
Is Discord safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 4 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Discord need?
It reads DISCORD_TOKEN from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Discord run?
It speaks stdio, so it runs as a local process your client starts. It is published on PyPI as mcp-discord.
How current is this page?
The grade is for one exact copy of the source (ced9701e70af), read on 2026-10-06. The repository is watched and re-audited when it changes.