Discourse MCP Server
for AI Agents

Q: What AI frameworks and AI clients does the StackOne Discourse MCP server support?

The StackOne Discourse MCP server supports both. MCP clients (paste-and-go apps): Claude Desktop, Claude Code, Cursor, VS Code, Goose. Agent frameworks (code SDKs you build with): OpenAI Agents SDK, Anthropic, Vercel AI, Google ADK, CrewAI, Pydantic AI, LangChain, LangGraph, Azure AI Foundry.

Production-ready Discourse MCP server with 67 extensible actions — plus built-in authentication, security, and optimized execution.

Start Free Book Demo Read Docs →

Discourse MCP Server

Built by

StackOne

Coverage

67 Agent Actions

Create, read, update, and delete across Discourse — and extend your agent's capabilities with custom actions.

MCP Actions → Create Actions →

Authentication

Agent Tool Authentication

Per-user OAuth in one call. Your Discourse MCP server gets session-scoped tokens with zero credentials stored on your infra.

Agent Auth →

Security

Agent Protection

Every Discourse tool response scanned for prompt injection in milliseconds — 88.7% accuracy, all running on CPU.

Prompt Injection Defense →

Performance

Max Agent Context. Min Cost.

Free up to 96% of your agent's context window to enhance reasoning and reduce cost, on every Discourse call.

Tools Discovery →

What is the Discourse MCP Server?

A Discourse MCP server lets AI agents read and write Discourse data through the Model Context Protocol — Anthropic's open standard for connecting LLMs to external tools. StackOne's Discourse MCP server ships with 67 pre-built actions, fully extensible via the Connector Builder — plus managed authentication, prompt injection defense, and optimized agent context. Connect it from MCP clients like Claude Desktop, Cursor, and VS Code, or from agent frameworks like OpenAI Agents SDK, LangChain, and Vercel AI SDK.

All Discourse MCP Tools and Actions

Every action from Discourse's API, ready for your agent. Create, read, update, and delete — scoped to exactly what you need.

Badges

Create Badge
Create a new custom badge (admin only).
List Badges
List all badges available in the Discourse instance (admin only).
Update Badge
Update the settings of an existing badge (admin only).
Delete Badge
Permanently delete a badge (admin only).

Categorys

Create Category
Create a new topic category (admin only).
Get Category
Retrieve a single category by its ID.
Update Category
Update the settings or metadata of an existing category (admin only).

Groups

Create Group
Create a new user group in Discourse (admin only).
List Groups
List all groups in the Discourse instance.
Get Group
Retrieve details about a specific group by its name.
Update Group
Update the settings or metadata of an existing group.
Delete Group
Permanently delete a group (admin only).

Group Members

Add Group Members
Add one or more users to a group.
List Group Members
Get all members of a specific group.
Remove Group Members
Remove one or more users from a group.

Posts

Create Post
Create a new reply in an existing topic, or create a new topic with an initial post.
Get Post
Retrieve a single post by its ID.
Update Post
Edit the Markdown content of an existing post.
Delete Post
Delete a post from a topic.

Private Messages

Create Private Message
Send a private message to one or more users or groups.
List Private Messages
Get private messages received by a specific user.

Tag Groups

Create Tag Group
Create a new tag group to organize related tags (admin only).
List Tag Groups
List all tag groups defined in the Discourse instance.
Get Tag Group
Retrieve a specific tag group by its ID.
Update Tag Group
Update the name or tags of an existing tag group (admin only).

Topics

Create Topic
Create a new topic in the Discourse forum.
Get Topic
Retrieve a single topic and its posts by topic ID.
Update Topic
Update the title, category, or tags of an existing topic.
Delete Topic
Delete a topic and all its posts (admin or moderator only).

Users

Create User
Create a new user account in the Discourse instance.
List Users
List all users in the Discourse instance (admin only).
Get User
Retrieve a user's public profile by their username.
Update User
Update a user's profile information.
Delete User
Permanently delete a user account (admin only).

Other (31)

Create Post Action
Perform an action on a post such as liking, bookmarking, or flagging.
Create Topic Timer
Schedule an automatic action on a topic at a future date and time.
List User Badges
Get all badges earned by a specific user.
List Categories
List all topic categories in the Discourse instance.
List Category Topics
Get all topics in a specific category.
Get Group By ID
Retrieve details about a specific group by its numeric ID.
List Notifications
List notifications for the currently authenticated user.
List Latest Posts
Get the latest posts from across all topics in the forum.
List Post Replies
Get all direct replies to a specific post.
List Sent Private Messages
Get private messages sent by a specific user.
Search
Search for topics, posts, users, and categories across the Discourse forum.
Get Site Info
Retrieve full site configuration and metadata for the Discourse instance.
Get Site Basic Info
Retrieve basic public information about the Discourse instance.
Get About
Retrieve public site statistics and staff information.
List Latest Topics
Get the latest topics from the Discourse forum, sorted by recent activity.
List Top Topics
Get the top-ranked topics filtered by time period from the Discourse forum.
Get Topic Posts
Retrieve specific posts from a topic by their post IDs.
List Users By Flag
List users filtered by account status flag (admin only).
Get User By ID
Retrieve a user's full details by their numeric ID (admin only).
List User Actions
Get a list of actions performed by a specific user.
Get User Emails
Retrieve the email addresses associated with a user account (admin only).
Get User By External ID
Retrieve a Discourse user by their external system ID (SSO/Discourse Connect lookup).
Update Topic Status
Enable or disable a status flag on a topic (close, pin, archive, hide, etc.).
Mark Notifications Read
Mark one or all notifications as read for the authenticated user.
Lock Post
Lock or unlock a post to prevent or allow editing.
Activate User
Activate a pending user account (admin only).
Deactivate User
Deactivate a user account to prevent them from logging in (admin only).
Suspend User
Suspend a user account until a specified date (admin only).
Log Out User
Force-log out a user's current session (admin only).
Silence User
Silence a user to prevent them from posting without fully suspending their account (admin only).
Anonymize User
Permanently anonymize a user by replacing all identifying information with random values (admin only).

Set Up Your Discourse MCP Server in Minutes

One endpoint. Any framework. Your agent is talking to Discourse in under 10 lines of code.

MCP Clients

Agent Frameworks

Claude Desktop-

JSON

Docs

{
  "mcpServers": {
    "stackone": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote@latest",
        "https://api.stackone.com/mcp?x-account-id=<account_id>",
        "--header",
        "Authorization: Basic <YOUR_BASE64_TOKEN>"
      ]
    }
  }
}

More Customer Support MCP Servers

123+ actions

104+ actions

84+ actions

74+ actions

71+ actions

69+ actions

Dixa

65+ actions

All Customer Support MCP Servers →

Discourse MCP Server Resources

MCP Code Mode: Keeping Tool Responses Out of Agent Context

Anthropic's code_execution processes data already in context. Custom MCP code mode keeps raw tool responses in a sandbox. 14K tokens vs 500.

11 min

Comparing BM25, TF-IDF, and Hybrid Search for MCP Tool Discovery

Benchmarking BM25, TF-IDF, and hybrid search for MCP tool discovery across 916 tools. The 80/20 TF-IDF/BM25 hybrid hits 21% Top-1 accuracy in under 1ms.

10 min

Indirect Prompt Injection Defense for MCP Tools: A Technical Guide

MCP tools that read emails, CRM records, and tickets are indirect prompt injection vectors. Here's how we built a two-tier defense that scans tool results in ~11ms.

12 min

Discourse MCP Server FAQ

Discourse MCP server vs direct API integration — what's the difference?

A Discourse MCP server and direct API integration serve different use cases. Direct API integration is for software-to-software — backend code calling Discourse. A Discourse MCP server is for AI agents — MCP clients like Claude and Cursor, plus framework agents built with OpenAI, LangChain, or Vercel AI — discovering and calling Discourse at runtime. StackOne provides both.

How does Discourse authentication work for AI agents?

Discourse authentication for AI agents works through a StackOne Connect Session. Create one via the dashboard or the SDK — you get an auth link and ready-to-paste config for Claude Desktop, Cursor, and other MCP clients. Your user authenticates their own Discourse account; StackOne handles token exchange, storage, and refresh. Credentials never reach the LLM, and each user is isolated via origin_owner_id.

Are Discourse MCP tools vulnerable to prompt injection?

Yes — Discourse MCP tools can be vulnerable to indirect prompt injection. Any tool that reads user-written content — documents, messages, tickets, records, or free-text fields — is a potential vector. StackOne Defender scans every tool response before it enters the agent's context — regex patterns in ~1ms, then a MiniLM classifier in ~4ms. 88.7% accuracy, CPU-only.

What is the context bloat of a Discourse agent and how do I avoid it?

Context bloat happens when Discourse tool schemas and API responses eat your Discourse agent's memory, preventing it from reasoning effectively. A single Discourse query can return a massive JSON response, and connecting multiple tools compounds the problem. Tools Discovery and Code Mode reduce context bloat — loading only relevant tools per query and keeping raw responses out of the agent's context.

Can I limit which actions my Discourse agent can access?

Yes — you can limit which actions your Discourse agent can access directly from the StackOne dashboard. Toggle actions on or off, or restrict them to specific accounts, with no code changes to your agent. Session tokens can be scoped to exact actions so if one leaks, exposure stays contained.

Can I create custom agent actions for my Discourse MCP server?

Yes — you can create custom agent actions for your Discourse MCP server using Connector Builder. It's an integration agent your coding assistant (Claude Code, Cursor, or Copilot) can invoke to research Discourse's API, generate production-ready connector YAML, test against the live API, and validate before you ship.

When should I NOT use a Discourse MCP server?

Skip a Discourse MCP server if your integration is purely software-to-software — direct Discourse API integration is simpler when no AI agent is involved. For deterministic, compliance-critical operations (financial transactions, regulatory reporting), direct API gives you predictable behavior without agent-driven decision-making. MCP shines when AI agents need to dynamically discover and call Discourse actions at runtime.

What AI frameworks and AI clients does the StackOne Discourse MCP server support?

The StackOne Discourse MCP server supports both. MCP clients (paste-and-go apps): Claude Desktop, Claude Code, Cursor, VS Code, Goose. Agent frameworks (code SDKs you build with): OpenAI Agents SDK, Anthropic, Vercel AI, Google ADK, CrewAI, Pydantic AI, LangChain, LangGraph, Azure AI Foundry.

Put your AI agents to work

All the tools you need to build and scale AI agent integrations, with best-in-class connectivity, execution, and security.

Start Free Book Demo

Discourse MCP Serverfor AI Agents