// SCAN REPORT

Discourse MCP

20 tools. 6 can modify or destroy data without limits.

1 destructive tool with no built-in limits. Policy required.

Last updated: 4 Jul 2026

6 can modify or destroy data

14 read-only

20 tools total

Community server · catalogue entry verified 04/07/2026

How to control Discourse MCP ↓

TOOL MAP

What Discourse MCP exposes to your agents

Read (14) Write / Execute (5) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Discourse MCP tools

Destructive · Medium discourse_delete_draft This tool irreversibly deletes data (draft content) and cannot be undone. Write · Medium discourse_create_category This tool creates a new category, which is a reversible write operation (categories can be deleted or modified later). Write · High discourse_create_user Creating a user account is a reversible write operation that adds a record to the system. Write · Medium discourse_create_post This tool creates new content (a post/reply) in a forum topic, which is a reversible write operation.

6 of Discourse MCP's 20 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Discourse MCP

PolicyLayer is an MCP gateway — it sits between your AI agents and Discourse MCP, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "discourse_delete_draft": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "discourse_create_category": {
    "limits": [
      {
        "counter": "discourse_create_category_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "discourse_get_chat_messages": {
    "limits": [
      {
        "counter": "discourse_get_chat_messages_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Discourse MCP — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON DISCOURSE →

Instant setup, no code required.

FULL CATALOGUE

All 20 Discourse MCP tools

DESTRUCTIVE 1 tools

Destructive discourse_delete_draft Delete a draft by its key. Requires the current sequence number from list/get operations to prevent conflicts.

WRITE 5 tools

Write discourse_create_category Create a new category in Discourse. Requires admin API key and write permissions. Rate limited to 1 request pe Write discourse_create_post Create a new post (reply) in an existing topic. Requires write permissions. Rate limited to 1 request per seco Write discourse_create_topic Create a new topic with the given title and first post. Requires write permissions. Rate limited to 1 request Write discourse_create_user Create a new user account in Discourse. Requires admin API key and write permissions. Write discourse_save_draft Create a draft topic, create a draft reply, or update an existing draft. Use this when the user wants to draft

READ 14 tools

Read discourse_get_chat_messages Get messages from a chat channel with flexible pagination and date-based filtering. Supports: (1) paginating w Read discourse_get_draft Retrieve a specific draft by its key. Common keys: Read discourse_get_user Get information about a Discourse user by username. Returns name, trust level, join date, bio, and profile lin Read discourse_list_categories List all categories visible to the current user. Returns category names and topic counts. Useful for discoveri Read discourse_list_chat_channels List all public chat channels visible to the current user. Returns channel information including title, descri Read discourse_list_drafts List all drafts for the current user. Returns draft keys, sequences, and preview content. Use this to find exi Read discourse_list_tags List all available tags on the Discourse site (if tagging is enabled). Returns tag names and usage counts. Use Read discourse_list_user_chat_channels List all chat channels for the currently authenticated user, including both public channels they Read discourse_list_user_posts Get a paginated list of posts and replies by a specific user, with the most recent first. Returns 30 posts per Read discourse_read_post Read a specific post by its ID. Returns the post content, author, creation date, and link to the post. Read discourse_read_topic Read a Discourse topic including its metadata (title, category, tags) and posts. Supports pagination for long Read discourse_search Search Discourse site content including topics, posts, and users. Use search operators like @username, tag, ca Read discourse_select_site Validate and select a Discourse site for subsequent tool calls. Verifies the site is reachable and retrieves i Read search_discourse_communities Discover Discourse forum communities by topic or find similar communities.

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Discourse MCP

Can an AI agent delete data through the Discourse MCP server? +

Yes. The Discourse MCP server exposes 1 destructive tools including discourse_delete_draft. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Discourse MCP? +

The Discourse MCP server has 5 write tools including discourse_create_category, discourse_create_post, discourse_create_topic. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Discourse MCP.

How many tools does the Discourse MCP server expose? +

20 tools across 3 categories: Destructive, Read, Write. 14 are read-only. 6 can modify, create, or delete data.

How do I enforce a policy on Discourse MCP? +

Register the Discourse MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Discourse MCP tool call.

Deterministic rules across all 20 Discourse MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON DISCOURSE →

Instant setup, no code required.

20 Discourse MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.