🚨 CVE-2026-53542
Termix is a web-based server management platform with SSH terminal, tunneling, and file editing capabilities. Prior to 2.3.2, the archive creation endpoint in src/backend/ssh/file-manager.ts passes selected file basenames to tar without an end-of-options marker and without making the operands unambiguously relative. A user with access to an SSH file-manager session can select basenames beginning with GNU tar options such as --checkpoint=1 and --checkpoint-action=exec, causing tar, tar.gz, tar.bz2, or tar.xz creation to interpret those names as options. The resulting checkpoint action executes commands on the managed SSH host with the privileges of the connected SSH account, allowing file disclosure, modification, and service disruption. This issue is fixed in version 2.3.2.

🎖@cveNotify

🚨 CVE-2026-53545
Termix is a web-based server management platform with SSH terminal, tunneling, and file editing capabilities. Prior to 2.3.2, the DELETE /ssh/tunnel/disconnect/:tunnelName teardown path in src/backend/ssh/tunnel.ts interpolates endpointPort, sourcePort, endpointUsername, and endpointIP into single-quoted pkill -f patterns. An authenticated user who can edit a tunnel host field can include a single quote to terminate the pattern and append a shell command, which executes when the tunnel is disconnected. Successful exploitation runs arbitrary commands on the source SSH host with the privileges of the connected SSH account. This issue is fixed in version 2.3.2.

🎖@cveNotify

🚨 CVE-2026-53546
Termix is a web-based server management platform with SSH terminal, tunneling, and file editing capabilities. Prior to 2.3.2, the terminal WebSocket accepts a user-controlled hostConfig.id and src/backend/ssh/host-resolver.ts resolves that host without requiring ownership or explicit access. When no credential is shared with the requester, resolveHostById performs an owner credential fallback, and src/backend/ssh/terminal.ts combines that credential with attacker-controlled ip, port, and username values. An authenticated low-privileged user can therefore make Termix authenticate to an attacker-controlled SSH server and disclose another user's stored SSH password or private-key material while the victim user's data key is unlocked. This issue is fixed in version 2.3.2.

🎖@cveNotify

🚨 CVE-2026-53547
Termix is a web-based server management platform with SSH terminal, tunneling, and file editing capabilities. Prior to 2.3.2, the POST /database/export endpoint creates a user export that includes the global settings table even though the rest of the export is user-scoped. The settings table contains reset_code_ and temp_reset_token_ password-reset artifacts, allowing a low-privileged authenticated user to recover another local account's reset code and complete the normal password-reset flow. Successful exploitation results in local-user account takeover and administrative compromise when the victim is an administrator. This issue is fixed in version 2.3.2.

🎖@cveNotify

🚨 CVE-2026-53548
Termix is a web-based server management platform with SSH terminal, tunneling, and file editing capabilities. Prior to 2.6.1, the GET /host/db/host/:id/password endpoint in src/backend/database/routes/host.ts accepts an authenticated user's numeric host ID and the field=password or field=sudoPassword query without enforcing host ownership during credential resolution. A failed requester-scoped lookup can resolve the host with the owner's context and return the owner's plaintext credential, allowing any authenticated user with a valid JWT to enumerate sequential hosts.id values and retrieve SSH or sudo passwords belonging to other users. The disclosed credentials can then be used to access and control managed systems outside the Termix instance. This issue is fixed in version 2.6.1.

🎖@cveNotify

🚨 CVE-2026-53549
Termix is a web-based server management platform with SSH terminal, tunneling, and file editing capabilities. Prior to 2.3.2, the POST /host/db/proxy/test endpoint accepts the singleProxy, proxyChain, and testTarget request fields without validating their destination addresses. The testProxyConnectivity path uses raw TCP and SOCKS connections to attacker-selected hosts and ports, allowing an authenticated user to probe localhost, private networks, link-local metadata services, and other infrastructure reachable from the Termix server. Structured connection errors disclose host reachability and timing information, and successful metadata access can expose cloud credentials. This issue is fixed in version 2.3.2.

🎖@cveNotify

🚨 CVE-2026-54491
Koel is a free, open-source music streaming solution. Prior to 9.7.1, outbound podcast and radio fetch paths perform a point-in-time App\Helpers\Network::isPublicHost() or isSafeUrl() check without pinning the validated address, and most paths lack redirect-hop validation and do not revalidate every redirect target. PhanAn\Poddle\Poddle::fromUrl(), PodcastService::getStreamableUrl(), PodcastService::isPodcastObsolete(), App\Rules\HasAudioContentType, and App\Rules\SafeUrl can therefore follow an attacker-controlled redirect to an internal address or connect after DNS rebinding changes a public resolution to a private one. These paths are reachable through podcast and radio APIs, including createPodcastChannel, createInternetRadioStation, refreshPodcasts, apiResource podcasts, and radio/stations, allowing an authenticated user to request internal services or cloud metadata and potentially receive parsed or streamed response content. This issue is fixed in version 9.7.1.

🎖@cveNotify

🚨 CVE-2026-54492
Koel is a free, open-source music streaming solution. Prior to 9.7.0, the Subsonic-compatible createPodcastChannel.view route accepts an authenticated user's private URL because app/Http/Requests/Subsonic/CreatePodcastChannelRequest.php does not apply the SafeUrl validation used by the regular podcast API. app/Http/Controllers/Subsonic/CreatePodcastChannelController.php passes the URL to app/Services/Podcast/PodcastService.php, where PodcastService::addPodcast() and createParser() invoke Poddle::fromUrl() during channel creation, causing immediate server-side requests to loopback, Docker bridge, or RFC1918 HTTP destinations. The confirmed impact is blind internal request execution because generic response-body exfiltration was not demonstrated through this route. This issue is fixed in version 9.7.0.

🎖@cveNotify

🚨 CVE-2026-54493
Koel is a free, open-source music streaming solution. Prior to 9.7.0, the Subsonic-compatible createInternetRadioStation.view and updateInternetRadioStation.view routes accept an authenticated user's streamUrl without the SafeUrl and HasAudioContentType checks used by the regular radio API. app/Http/Requests/Subsonic/CreateInternetRadioStationRequest.php and app/Http/Requests/Subsonic/UpdateInternetRadioStationRequest.php pass the stored URL through app/Services/RadioService.php to app/Services/Radio/RadioStreamProxy.php, where RadioStreamProxy::openStream() calls fopen($url, 'r', false, $context). Streaming /radio/stream/{id} returns the upstream response body, allowing access to loopback, RFC1918, Docker bridge, metadata, or other internal HTTP services reachable from the Koel server. This issue is fixed in version 9.7.0.

🎖@cveNotify

🚨 CVE-2026-54494
Koel is a free, open-source music streaming solution. Prior to 9.7.1, App\Helpers\Network::isPublicHost() uses filter_var() with FILTER_FLAG_NO_PRIV_RANGE and FILTER_FLAG_NO_RES_RANGE, which treats NAT64 64:ff9b::/96 and 6to4 2002::/16 wrappers of private, loopback, or link-local IPv4 addresses as public. An authenticated user can place such an address in a podcast RSS <enclosure url> and reach app/Values/Podcast/EpisodePlayable.php through EpisodePlayable::createForEpisode(), where isSafeUrl() accepts the target and Http::sink($file)->get($url) fetches it. On a host with NAT64 or 6to4 routing, Koel can request internal services or cloud metadata and return the response body to the user. This issue is fixed in version 9.7.1

🎖@cveNotify

🚨 CVE-2026-54738
Lemmy is a link aggregator and forum for the fediverse. Prior to 0.19.19 and 1.0.0-beta.1, actix-web ConnectionInfo::realip_remote_addr reads the first value of X-Forwarded-For as the client address used by raw_ip_key in crates/utils/src/rate_limit/mod.rs. Lemmy's bundled docker/nginx.conf uses $proxy_add_x_forwarded_for instead of $remote_addr, which appends the real client address to an X-Forwarded-For value supplied by the client. An unauthenticated attacker can therefore place a different spoofed address first on each request and receive a new rate-limit bucket, bypassing limits on POST /api/v4/account/auth/register, POST /api/v4/account/auth/login, POST /api/v4/post, POST /api/v4/comment, GET /api/v4/search, POST /api/v4/image, and POST /api/v4/account/import_settings. This permits excessive account creation, brute-force attempts, spam, scraping, uploads, and repeated imports. This issue is fixed in versions 0.19.19 and 1.0.0-beta.1.

🎖@cveNotify

🚨 CVE-2026-54739
Lemmy is a link aggregator and forum for the fediverse. Prior to 0.19.19 and 1.0.0-beta.1, Lemmy's login endpoint in crates/api/api/src/local_user/login.rs returns different errors depending on whether the username_or_email value exists. LocalUserView::find_by_email_or_name propagates a NotFound response for an unknown account, while an existing account with a wrong password returns LemmyErrorType::IncorrectLogin. This observable response discrepancy, including HTTP 404 for a nonexistent account and HTTP 400 for an incorrect password on an existing account, allows an unauthenticated attacker to confirm registered usernames or email addresses and use the results for targeted credential attacks or social engineering. This issue is fixed in versions 0.19.19 and 1.0.0-beta.1.

🎖@cveNotify

🚨 CVE-2026-54740
Lemmy is a link aggregator and forum for the fediverse. Prior to 0.19.19 and 1.0.0-alpha.18, a lower-ranked remote moderator can remove a higher-ranked moderator by sending a signed ActivityPub Remove activity to the target instance. The local API uses LocalUser::is_higher_mod_or_admin_check to enforce moderator rank, but CollectionRemove::verify in crates/apub/activities/src/community/collection_remove.rs only calls verify_mod_action. CollectionRemove::receive dereferences self.object as an ApubPerson, creates a CommunityModeratorForm, and calls CommunityActions::leave without checking that the actor outranks the moderator identified by the object field. In communities with federated moderators, a junior moderator can therefore strip senior moderators from the community moderator list even though the local API rejects the same action. This issue is fixed in versions 0.19.19 and 1.0.0-alpha.18.

🎖@cveNotify

🚨 CVE-2026-54741
Lemmy is a link aggregator and forum for the fediverse. Prior to 0.19.19 and 1.0.0-alpha.18, Lemmy blocks new private messages from a sender after the recipient blocks that sender, but the edit path skips the same block check. create_private_message checks the recipient's block list with PersonActions::read_block before inserting a message, while edit_private_message in crates/api/api_crud/src/private_message/update.rs only checks that the caller is orig_private_message.creator_id. The update then writes new content and returns the modified PrivateMessageView without consulting the recipient's block list, allowing a blocked sender to keep changing an old message that the recipient can still see and providing a post-block harassment path. This issue is fixed in versions 0.19.19 and 1.0.0-alpha.18.

🎖@cveNotify

🚨 CVE-2026-54743
Lemmy is a link aggregator and forum for the fediverse. Prior to lemmy-ui 0.19.19-beta.1, LemmyNet/lemmy-ui renders Markdown in src/shared/markdown.ts for post bodies, comment bodies, private messages, and community and site sidebars through mdToHtml, which returns a raw __html object that Inferno injects without a sanitizer pass. setupMarkdown configures html as false but registers markdown-it-html5-embed@1.0.0 with useImageSyntax enabled, so an image targeting video media becomes a video element whose fallback incorporates the image alt text through unescaped string replacement. The html setting does not apply to plugin-generated output, allowing crafted alt text to reach the DOM as live HTML in contexts that do not use mdToHtmlNoImages. An approved member or a remote federated instance can store such content, and a viewer who renders it may execute JavaScript in the lemmy-ui origin, exposing the viewer's session and authenticated actions. The advisory notes that Content Security Policy prevents the described exploit in production, but also states that the tested default self-hosted deployment serves no Content-Security-Policy. This issue is fixed in lemmy-ui version 0.19.19-beta.1.

🎖@cveNotify

🚨 CVE-2026-61556
LiquidJS is a Shopify / GitHub Pages compatible template engine in pure JavaScript. From 10.26.0 until 10.27.1, the strip_html filter in src/filters/html.ts can enter an infinite loop when an input string contains <, includes at least one preceding character, and has no later >. In strip_html, the search for the next opener advances lt while the loop index remains unchanged when the closer search returns -1, and the equality-only stall guard does not exit because the loop index is less than lt. Reprocessing the same state indefinitely blocks template rendering and can cause denial of service with an input as short as a<. This issue is fixed in version 10.27.1.

🎖@cveNotify

🚨 CVE-2026-62727
Concurrent execution using shared resource with improper synchronization ('race condition') in Windows Telephony Service allows an authorized attacker to elevate privileges locally.

🎖@cveNotify

🚨 CVE-2026-68553
Coturn is a free open source implementation of TURN and STUN Server. Prior to 4.13.0, an authenticated TURN user can place printf-style format specifiers in the STUN USERNAME or REALM attribute, which passes is_secure_string() validation and is embedded into Redis keys at nine call sites in src/apps/relay/ns_ioalib_engine_impl.c. send_message_to_redis() in src/apps/relay/hiredis_libevent2.c then passes the attacker-controlled key as the format argument to redisAsyncCommand() while supplying only one variadic value, causing hiredis redisvFormatCommand() to read past the va_list. Exploitation can crash the coturn process and terminate active TURN sessions or disclose stack memory into Redis. This issue is fixed in version 4.13.0.

🎖@cveNotify

🚨 CVE-2026-68554
Coturn is a free open source implementation of TURN and STUN Server. Prior to 4.15.0, an on-path attacker can append attributes after MESSAGE-INTEGRITY to an authenticated STUN request on plain UDP or TCP, adjust the STUN header length, and recompute the unkeyed FINGERPRINT while the original HMAC remains valid because it covers only the message prefix. Server-side parsing in src/server/ns_turn_server.c continues past MESSAGE-INTEGRITY through handle_turn_allocate(), handle_turn_create_permission(), handle_turn_refresh(), and handle_turn_command(), allowing trailing LIFETIME, XOR-PEER-ADDRESS, or ORIGIN attributes to override allocation lifetime, inject a permission, or bypass the origin check. TLS and DTLS deployments prevent this in-transit modification. This issue is fixed in version 4.15.0.

🎖@cveNotify

🚨 CVE-2026-68555
Coturn is a free open source implementation of TURN and STUN Server. In 4.15.0, an authenticated TURN user can repeatedly resume one allocation from fresh UDP 5-tuples without completing a handoff when the server enables --mobility. mobile_begin_transition() in src/server/ns_turn_server.c disarms each new session's allocation timeout and overwrites the allocation's single mobile_pending_resume link, leaving earlier pending sessions unreachable by the cleanup path, while copy_auth_parameters() ignores inc_quota() failure. The attacker can therefore retain unbounded server-side sessions and exhaust process memory even when --user-quota=1 is configured. This issue is fixed in version 4.16.0.

🎖@cveNotify

🚨 CVE-2026-69222
LiquidJS is a Shopify / GitHub Pages compatible template engine in pure JavaScript. Prior to 10.27.2, the join filter in src/filters/array.ts computes complexity from array.length and separator length instead of the total string length produced by array.join(sep). The concat filter can cheaply double arrays of references, after which join materializes the referenced content while charging only for element count, allowing a template to exceed a configured memoryLimit by a large factor. The sibling array_to_sentence_string filter in src/filters/string.ts has the same accounting defect, and a crafted template can allocate toward V8's string or process memory limit and crash the process. This issue is fixed in version 10.27.2.

🎖@cveNotify