
Company News
Free Business Plan Upgrades for Open Source Maintainers
Open source maintainers are under more pressure than ever. We're raising our open source program from the Team plan to the Business plan, free.
openclaw-discourse
Advanced tools
OpenClaw plugin for Discourse forum integration — read, search, filter, and write topics/posts via agent tools
OpenClaw plugin for Discourse forum integration. Read, search, and filter topics and posts. Optionally create and update content with an API key.
openclaw plugins install openclaw-discourse
Or install from GitHub:
openclaw plugins install github:pranciskus/discourse-openclaw
openclaw plugins update openclaw-discourse
After installing, configure the plugin with your forum URL:
openclaw config set plugins.entries.openclaw-discourse.config.siteUrl "https://meta.discourse.org"
For authenticated access (private categories, write tools):
openclaw config set plugins.entries.openclaw-discourse.config.apiKey "your-discourse-api-key"
openclaw config set plugins.entries.openclaw-discourse.config.allowWrites true
Or edit openclaw.json directly:
{
"plugins": {
"entries": {
"openclaw-discourse": {
"config": {
"siteUrl": "https://meta.discourse.org"
}
}
}
}
}
{
"plugins": {
"entries": {
"openclaw-discourse": {
"config": {
"siteUrl": "https://community.example.com",
"apiKey": "your-discourse-api-key",
"apiUsername": "system",
"staffUsernames": ["admin", "moderator1"],
"categories": ["support", "general"],
"allowWrites": true,
"signature": "*This content was written by AI.*",
"requestTimeoutMs": 15000
}
}
}
}
}
| Option | Type | Required | Default | Description |
|---|---|---|---|---|
siteUrl | string | Yes | — | Discourse forum base URL |
apiKey | string | No | — | API key for auth. Required for write tools and private categories |
apiUsername | string | No | "system" | Discourse user to act as when using a global API key. Posts and actions will appear under this username. |
staffUsernames | string[] | No | [] | Usernames treated as staff in discourse_unanswered |
categories | string[] | No | [] | Category slugs to scope monitoring (empty = all) |
allowWrites | boolean | No | false | Enable write tools. Requires apiKey |
signature | string | No | *This content was written by AI.* | Appended to all AI-generated posts and topics |
requestTimeoutMs | number | No | 15000 | HTTP request timeout in ms |
| Tool | Description | Key Parameters |
|---|---|---|
discourse_read_topic | Read a topic by ID with posts | topic_id, post_limit |
discourse_read_post | Read a single post by ID | post_id |
discourse_search | Full-text search across topics and posts | query, max_results |
discourse_filter_topics | List/filter recent topics by category | category_slug, order, page |
discourse_unanswered | Find recent topics with no staff replies | hours, category_slug |
discourse_get_user | Get a user's profile | username |
discourse_list_user_posts | List a user's recent posts | username, offset, limit |
discourse_get_categories | List all forum categories | — |
discourse_get_tags | List all tags with counts | — |
discourse_site_rules | Fetch the site's AI usage policy (/llms.txt) | — |
apiKey + allowWrites: true)| Tool | Description | Key Parameters |
|---|---|---|
discourse_create_post | Reply to an existing topic | topic_id, raw |
discourse_create_topic | Create a new topic | title, raw, category_id, tags |
discourse_update_topic | Update topic title/category/tags | topic_id, title, category_id, tags |
Write tools are rate-limited to ~1 request/second.
The plugin respects the llms.txt convention. The discourse_site_rules tool fetches /llms.txt from the forum on demand and returns the site's AI usage policy. All write tool descriptions instruct the LLM to call discourse_site_rules first and follow the rules before posting.
The result is cached for 24 hours. If the fetch fails due to a network error, the LLM is told to proceed with caution rather than assuming no restrictions.
All AI-generated posts and topics automatically include a signature separated by a horizontal rule (---). This applies to create_post and create_topic only — update_topic changes metadata (title/category/tags), not body content. The signature text is configurable via the signature config field and is appended programmatically — the LLM cannot skip it.
https://your-forum.com/admin/api/keys| Feature | discourse-openclaw | discourse-mcp |
|---|---|---|
| Platform | OpenClaw native | MCP (Claude Desktop, etc.) |
| Read tools | 10 | 10 |
| Write tools | 3 (focused) | 9 (full admin) |
| Unique tools | discourse_unanswered | Chat, drafts, admin tools |
| Install | openclaw plugins install | Standalone CLI |
| Best for | OpenClaw agents, community support workflows | General MCP clients |
This plugin is designed for agent-driven community support. The discourse_unanswered tool is specifically built for finding topics that need attention, which is not available in discourse-mcp.
git clone https://github.com/pranciskus/discourse-openclaw
cd discourse-openclaw
npm install
npm test
npm run typecheck
MIT
FAQs
OpenClaw plugin for Discourse forum integration — read, search, filter, and write topics/posts via agent tools
The npm package openclaw-discourse receives a total of 0 weekly downloads. As such, openclaw-discourse popularity was classified as not popular.
We found that openclaw-discourse demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Company News
Open source maintainers are under more pressure than ever. We're raising our open source program from the Team plan to the Business plan, free.

Security News
The supply chain control that delays freshly published gems now covers lockfile generation and gem vendoring in Ruby projects.

Security News
During a UK cyber test, a Mythos 5 agent used sockpuppets, social engineering, and prompt injection to try to get a maintainer to merge malware.