Discourse is an open source discussion platform. Versions prior to 2025.12.2, 2026.1.1, and 2026.2.0 have an IDOR (Insecure Direct Object Reference) in `ReviewableNotesController`. When `enable_category_group_moderation` is enabled, a user belonging to a category moderation group can create or delete their own notes on **any** reviewable in the system, including reviewables in categories they do not moderate. The controller used an unscoped `Reviewable.find` and the `ensure_can_see` guard only checked whether the user could access the review queue in general, not whether they could access the specific reviewable. Only instances with `enable_category_group_moderation` enabled are affected. Staff users (admins/moderators) are not impacted as they already have access to all reviewables. The issue is patched in versions 2025.12.2, 2026.1.1, and 2026.2.0 by scoping the reviewable lookup through `Reviewable.viewable_by(current_user)`. As a workaround, disable the `enable_category_group_moderation` site setting. This removes the attack surface as only staff users will have access to the review queue.

Discourse is an open source discussion platform. Prior to versions 2025.12.2, 2026.1.1, and 2026.2.0, when the `patreon_webhook_secret` site setting is blank, an attacker can forge valid webhook signatures by computing an HMAC-MD5 with an empty string as the key. Since the request body is known to the sender, the attacker can produce a matching signature and send arbitrary webhook payloads. This allows unauthorized creation, modification, or deletion of Patreon pledge data and triggering patron-to-group synchronization. This vulnerability is patched in versions 2025.12.2, 2026.1.1, and 2026.2.0. The fix rejects webhook requests when the webhook secret is not configured, preventing signature forgery with an empty key. As a workaround, configure the `patreon_webhook_secret` site setting with a strong, non-empty secret value. When the secret is non-empty, an attacker cannot forge valid signatures without knowing the secret.

Discourse is an open source discussion platform. In versions prior to 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0, authenticated users can submit crafted payloads to /drafts.json that cause O(n^2) processing in Base62.decode, tying up workers for 35-60 seconds per request. This affects all users as the shared worker pool becomes exhausted. This issue is patched in versions 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0. Lowering the max_draft_length site setting reduces attack surface but does not fully mitigate the issue, as payloads under the limit can still trigger the slow code path.

Discourse is an open source discussion platform. In versions prior to 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0, non-admin moderators with the `moderators_change_post_ownership` setting enabled can change ownership of posts in private messages and restricted categories they cannot access, then export their data to view the content. This is a broken access control vulnerability affecting sites that grant moderators post ownership transfer permissions. This issue is patched in versions 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0. The patch adds visibility checks for both the topic and posts before allowing ownership transfer. As a workaround, disable the `moderators_change_post_ownership` site setting to prevent non-admin moderators from using the post ownership transfer feature.

Discourse is an open source discussion platform. In versions prior to 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0, an endpoint lets any authenticated user bypass the ai_discover_persona access controls and gain ongoing DM access to personas that may be wired to staff-only categories, RAG document sets, or automated tooling, enabling unauthorized data disclosure. Because the controller also accepts arbitrary user_id, an attacker can impersonate other accounts to trigger unwanted AI conversations on their behalf, generating confusing or abusive PM traffic. This issue is patched in versions 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0. No known workarounds are available.

Discourse is an open source discussion platform. In versions prior to 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0, moderators can access the `top_uploads` admin report which should be restricted to admins only. This report displays direct URLs to all uploaded files on the site, including sensitive content such as user data exports, admin backups, and other private attachments that moderators should not have access to. This issue is patched in versions 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0. There is no workaround. Limit moderator privileges to trusted users until the patch is applied.

Discourse is an open source discussion platform. Versions prior to 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0 have a content-security-policy-mitigated cross-site scriptinv vulnerability on the Discourse Math plugin when using its KaTeX variant. This issue is patched in versions 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0. As a workaround, the Discourse Math plugin can be disabled, or the Mathjax provider can be used instead of KaTeX.

Discourse is an open source discussion platform. In versions prior to 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0, a hostname validation issue in FinalDestination could allow bypassing SSRF protections under certain conditions. This issue is patched in versions 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0. No known workarounds are available.

Discourse is an open source discussion platform. Versions prior to 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0 have an application level denial of service vulnerabilityin the username change functionality at try.discourse.org. The vulnerability allows attackers to cause noticeable server delays and resource exhaustion by sending large JSON payloads to the username preference endpoint PUT /u//preferences/username, resulting in degraded performance for other users and endpoints. This issue is patched in versions 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0. No known workarounds are available.

Discourse is an open source discussion platform. In versions prior to 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0, non-admin moderators can view sensitive information in staff action logs that should be restricted to administrators only. The exposed information includes webhook payload URLs and secrets, API key details, site setting changes, private message content, restricted category names and structures, and private chat channel titles. This allows moderators to bypass intended access controls and extract confidential data by monitoring the staff action logs. With leaked webhook secrets, an attacker could potentially spoof webhook events to integrated services. This issue is patched in versions 3.5.4, 2025.11.2, 2025.12.1, and 2026.1.0. As a workaround, site administrators should review and limit moderator appointments to fully trusted users. There is no configuration-based workaround to prevent this access.