CVE Watch402,460 in full archive

Vulnerabilities exploitable today

16,759in current view

Single score combining CVSS, KEV membership and EPSS. Every CVE with its own record — timeline from publication to active exploitation.

In KEV catalog1,754

New KEV · 24H0

Exploit Today ≥ 701,668

Distribution · last window

Critical
3,959
High
16,991
Medium
15,819
Low
1,541

Filters

Window

Severity

Flags

Vulnerabilities8,921–8,960 · 16,759

CVECVSSEPSSKEVRExploitTitleMod.

CVE-2026-1095687.1 HIG

15.0%

——4Shaark 1.2.13 through 1.2.44 contains a missing authorization vulnerability in the Link, Story, Chest and Album API controllers that skips ownership checks. Authenticated non-admin users can send PUT or DELETE requests to /api/{link,story,chest,album}/{id} with numeric IDs to overwrite or delete other users' content, including private items.7d CVE-2026-1095677.2 HIG

24.8%

——7Rapyd Admin from 9.0.0 before 9.20.0 contains an improper privilege management vulnerability in the UsersEdit Livewire component that allows users with the edit users permission to assign themselves the admin role. Attackers holding roles such as the shipped operator role can submit the admin role ID for their own account, lifting tenant scoping, bypassing per-record authorization, and enabling impersonation.7d CVE-2026-1095669.8 CRI

37.2%

——11Rapyd Admin from 9.0.0 before 9.20.0 contains a hard-coded credentials vulnerability that allows remote unauthenticated attackers to gain super-admin access through an AuthSeeder-created admin@laravel account with password admin. The account is seeded by the setup installer and silently re-created on every rpd:make run, letting attackers log in as a global super admin bypassing per-record authorization and tenant scoping.7d CVE-2026-1095656.1 MED

10.8%

——3Rapyd Admin from 9.4.0 before 9.20.0 contains a stored cross-site scripting vulnerability that allows unauthenticated attackers to inject scripts through self-registered display names. Attackers can register with an HTML or JavaScript name rendered unescaped in the Activity log Author filter, executing in an administrator's session.7d CVE-2026-1095643.1 LOW

8.2%

——2LearnHouse before 1.4.0 contains a missing authorization vulnerability in POST /api/v1/playgrounds/{uuid}/duplicate that skips the read-access and draft checks enforced by get_playground. Authenticated users with playground creation rights, such as Instructors, who know a RESTRICTED or unpublished playground UUID can copy its html_content and republish it publicly.7d CVE-2026-1095637.5 HIG

27.5%

——8N3.js before 2.11.1 contains a regular expression denial of service vulnerability in the N3Lexer number pattern that backtracks quadratically on long digit runs. Attackers can supply untrusted Turtle or N3 input containing long runs of digits that do not end a number, blocking the Node.js event loop and causing denial of service.7d CVE-2026-1095625.4 MED

11.7%

——4SnapOtter 1.15.10 through 2.2.0 contains a missing authorization vulnerability that allows any authenticated principal to call the /api/v1/upload, /api/v1/preview and /api/v1/preview/generate handlers without permission checks. Attackers holding even a settings:read-scoped API key can stage arbitrary files outside storage quota, served unauthenticated via /api/v1/download, and feed crafted input to image and video decoders.7d CVE-2026-1095614.2 MED

8.7%

——3goose from 1.28.0 before 1.44.0 contains an authorization bypass vulnerability that allows tools to run without user approval in Manual Approval mode because persisted read-only annotations are never cleared by update_goose_mode. After a session uses Smart Approval mode, prompt-injected calls to tools annotated readOnlyHint: true execute without prompting, bypassing the user's selected approval mode in PermissionInspector::inspect.7d CVE-2026-1095607.5 HIG

32.5%

——10Continue CLI (@continuedev/cli) through 1.5.47 contains a permission bypass vulnerability that allows attackers to run unapproved commands by chaining them after allowed Bash prefix rules. Attackers can inject instructions into repository files or fetched pages so the agent appends commands after ;, && or | to execute code as the user.7d CVE-2026-1095598.7 HIG

22.5%

——7EdgeEver Desktop 1.6.26 through 1.112.0 contains a navigation allowlist bypass that lets attackers load arbitrary origins in the main window via string-prefix URL matching. Attackers can embed a link using userinfo syntax after the dev-server prefix, for example in a note-template preview, to obtain the edgeeverDesktop bridge including session token, sidecar RPC, and ACP channels.7d CVE-2026-1095588.7 HIG

24.9%

——7EdgeEver 1.6.0 through 1.112.0 contains a stored cross-site scripting vulnerability in TemplatesPane.tsx that renders template contentMarkdown with raw HTML via dangerouslySetInnerHTML without sanitization. Attackers can write templates containing HTML event handlers through an API or MCP token to run script on preview and access the privileged window.edgeeverDesktop bridge.7d CVE-2026-1095578.7 HIG

24.9%

——7EdgeEver through 1.112.0 contains a stored cross-site scripting vulnerability in setMemoDragPreview that allows attackers to inject script through notebook names or memo titles. Attackers can set malicious names via MCP create_notebook, rename_notebook or create_memo tools, running script on drag with access to the desktop bridge's session token.7d CVE-2026-1095568.0 HIG

42.9%

——13EdgeEver Desktop 1.92.0 through 1.112.0 contains a remote code execution vulnerability in the desktop:acp-probe IPC handler, which lacks a sender check and spawns a caller-supplied input.path executable. Attackers with script execution in a renderer, such as via template-preview or drag-preview stored cross-site scripting, can call probeAcpAdapter to launch local executables with the user's privileges.7d CVE-2026-1094276.5 MED

10.4%

——3Telegram-Stremio through 5.1.0 contains a broken object-level authorization vulnerability that allows authenticated token holders to access arbitrary chat messages by forging unsigned, unauthenticated stream IDs decoded via encode_string/decode_string functions. Attackers can craft or enumerate sequential msg_id values and set a global flag in the stream_handler endpoint to route requests through the operator's personal Userbot account, exposing private channels, groups, and Saved Messages, while the unauthenticated thumb_handler endpoint further discloses thumbnails from those same chats without any token validation.7d CVE-2026-1094267.5 HIG

1.2%

——0Telegram-Stremio through 5.1.0 contains an insecure storage vulnerability that allows attackers with access to the MongoDB database, backups, or logs to recover the operator's authenticated Telegram session string, which is protected only by a keyless, reversible encoding transform rather than actual encryption. Attackers can round-trip the stored value through the repository's own decode_string function with no secret key to obtain a fully authenticated Telegram session, granting complete control of the operator's Telegram account even when two-step verification is enabled.7d CVE-2026-1093807.5 HIG

37.8%

——11modern-tar 0.3.5 through 0.8.5 contains a memory exhaustion vulnerability that allows attackers to produce negative entry sizes via non-octal bytes parsed by readOctal(). Attackers can supply a crafted, optionally gzip-compressed, archive to unpackTar() that buffers chunks indefinitely, exhausting process memory and suppressing later entries.7d CVE-2026-1093795.3 MED

27.2%

——8Cockpit CMS 2.14.1 contains an observable timing discrepancy vulnerability that allows unauthenticated attackers to enumerate accounts due to an incomplete timing fix in /auth/check. Attackers can post login attempts and measure response times, since existing accounts incur two bcrypt operations versus one, revealing valid usernames and email addresses.7d CVE-2026-1093784.3 MED

14.6%

——4SPIP before 4.4.18 contains a server-side request forgery vulnerability in recuperer_url() in ecrire/inc/distant.php that follows up to 10 redirects without re-validating targets with valider_url_distante(). Authenticated attackers can supply a remote URL, such as a syndicated site feed or remote document, that redirects to internal or link-local addresses including cloud metadata endpoints.10d CVE-2026-1093774.3 MED

15.5%

——5ERPNext 9.0.0 through 16.50.0 contains a missing authorization vulnerability that allows authenticated users to view stock quantities via whitelisted batch methods. Attackers, including customer portal users, can call get_batch_qty, get_batches_by_oldest or get_pos_reserved_batch_qty to list every item batch with per-warehouse quantities.9d CVE-2026-1093764.3 MED

15.5%

——5ERPNext 15.55.0 through 16.50.0 contains an incorrect authorization vulnerability in the get_timesheet_html method that lets authenticated portal users view project Timesheets. The method only checks Project read access, ignoring hide_timesheets and project membership, so attackers can retrieve timesheet names, status, time-log dates and last modifiers.11d CVE-2026-1093756.1 MED

24.1%

——7Vercel deepsec from 0.1.0 before 1.1.17 contains a link following vulnerability in extractTarballLocally that allows sandboxed code to write files outside the sandbox by planting links in result tarballs. Attackers controlling sandbox code, such as a prompt-injected coding agent, can plant symbolic or hard links in the host data directory to write arbitrary files on the orchestrator host.7d CVE-2026-1089314.3 MED

17.2%

——5A vulnerability was found in Netcore NR268 1.7.121109. This affects the function eval of the file dhcp_dynamicip_show.cgi of the component DHCP Handler. The manipulation of the argument Hostname results in cross site scripting. The attack may be launched remotely. The exploit has been made public and could be used. The vendor was contacted early about this disclosure but did not respond in any way.9d CVE-2026-1089302.7 LOW

19.1%

——6A vulnerability has been found in Netcore NR268 1.7.121109. The impacted element is the function ddns_wan_list_show of the file ddns_wan_list_show.cgi of the component DDNS Handler. The manipulation of the argument user_idN/passwordN leads to information disclosure. The attack may be initiated remotely. The exploit has been disclosed to the public and may be used. The vendor was contacted early about this disclosure but did not respond in any way.10d CVE-2026-1089299.9 CRI

36.1%

——11A flaw has been found in Netcore NR268 1.7.121109. The affected element is the function execvp of the file /router/DDNSset.cgi of the component DDNS. Executing a manipulation of the argument Hostname can lead to argument injection. The attack can be launched remotely.7d CVE-2026-1089286.3 MED

15.4%

——5A security vulnerability has been detected in itsourcecode Online Admission System 1.0. This issue affects some unknown processing of the file /admin/sum.php. Such manipulation of the argument ID leads to sql injection. It is possible to launch the attack remotely. The exploit has been disclosed publicly and may be used.10d CVE-2026-1089276.3 MED

15.4%

——5A weakness has been identified in itsourcecode Baptism Information Management System 1.0. This vulnerability affects unknown code of the file pbaptism.php. This manipulation of the argument b_id causes sql injection. It is possible to initiate the attack remotely. The exploit has been made available to the public and could be used for attacks.9d CVE-2026-1064954.3 MED

14.1%

——4Backstage is an open framework for building developer portals. Prior to 0.17.8, the @backstage/backend-defaults package is affected by improper access control in backend action invocation. An authenticated user could invoke actions that an operator had excluded through action configuration. Existing action permissions and downstream authorization remain enforced. This issue is fixed in version 0.17.8.7d CVE-2026-1050527.5 HIG

49.2%

——15IBM Maximo Application Suite 8.10, 8.11, 9.0, 9.1, and 9.2 could allow a remote attacker to traverse directories on the system. An attacker could send a specially crafted URL request containing "dot dot" sequences (/../) to view arbitrary files on the system.10d CVE-2025-714299.8 CRI

86.7%

——26GPT-SoVITS through 20250606v2pro contains an OS command injection vulnerability in tools/uvr5/vr.py that allows unauthenticated attackers to execute commands via output folder paths. Attackers can enter shell substitutions like $(...) in the UVR5 WebUI output folder textboxes with mp3 or m4a export to run commands as the server user.4d CVE-2024-583896.1 MED

19.2%

——6HoloViz Panel from 1.2.2 before 1.9.0 contains an open redirect vulnerability in panel/auth.py that allows remote attackers to redirect users by supplying an unvalidated next query parameter. BasicLoginHandler and PasswordLoginHandler store the parameter in a next_url cookie, so crafted login links send victims to attacker-controlled external sites after successful login.7d CVE-2022-510205.9 MED

3.4%

——1Flower before 2.2.0 contains a cross-site request forgery vulnerability because Tornado xsrf_cookies is disabled and the OAuth session cookie lacks SameSite. Attackers can lure authenticated users to malicious pages that POST to /api/worker/shutdown, /api/worker/pool/restart, or /api/task/revoke to shut down Celery workers, resize pools, and revoke tasks.7d CVE-2021-480128.2 HIG

34.7%

——10fast-toml through 0.5.4 contains a prototype pollution vulnerability that allows attackers to modify Object.prototype by supplying TOML containing __proto__, constructor or prototype key segments. Attackers can pass [__proto__] or [constructor.prototype] table headers or __proto__.x dotted keys to parse() to pollute all objects, causing denial of service or authorization bypass.7d CVE-2021-480117.5 HIG

45.9%

——14Servor 1.0.0 through 4.0.2 contains a path traversal vulnerability that allows unauthenticated remote attackers to escape the served root by sending URL-encoded dot segments. Attackers can send %2e%2e/ sequences, which decodeURI() decodes before path.join() in serveStaticFile and serveDirectoryListing, to read arbitrary files and list directories outside the root.7d CVE-2020-372807.5 HIG

44.5%

——13cbor-sync through 1.0.4 contains a resource exhaustion vulnerability in CBOR.decode that trusts declared array and map lengths without checking remaining input. Remote attackers can supply a tiny crafted CBOR payload declaring huge lengths to drive billions of iterations, exhausting CPU and process memory.7d CVE-2020-372798.1 HIG

74.0%

——22easy-pdf-merge 0.1.0 through 0.2.6 contains an OS command injection vulnerability that allows attackers to execute commands by injecting shell metacharacters into source paths, destination path, or maxHeap option. Attackers can supply file names or options containing double quotes, semicolons, $() or backticks, which child_process.exec() runs through a shell with the Node.js process privileges.7d CVE-2026-960129.9 CRI

42.6%

——13IBM Db2 Mirror for i 7.4, 7.5, and 7.6 is vulnerable to privilege escalation. An authenticated attacker could craft a malicious request that bypasses additional authorization and input validation checks to execute malicious code. An attacker could exploit this vulnerability to obtain elevated privileges on the system.10d CVE-2026-95838—

11.2%

——3containerd is an open-source container runtime. Prior to 1.7.35, 2.0.12, 2.2.8, and 2.3.5, containerd's layer unpack handler repeatedly traverses directories for certain opaque whiteout configurations. A malicious image layer can trigger repeated directory traversals while opaque whiteouts are processed. Pulling or importing the image can consume CPU, delay container startup, and reduce image-pull concurrency for unrelated deployments. This issue is fixed in versions 1.7.35, 2.0.12, 2.2.8, and 2.3.5.7d CVE-2026-958377.8 HIG

1.1%

——0containerd is an open-source container runtime. From 2.1.0 until 2.2.7 and 2.3.4, containerd's CRI CreateContainer checkpoint restore is vulnerable because CRIU restores process security state from checkpoint data instead of enforcing the destination ContainerConfig. An attacker who can create a container from a crafted checkpoint can supply restored credentials, capabilities, no_new_privs, and seccomp state. The restored process can run as root with full capabilities and no enforced seccomp filters while CRI status reports the requested configuration. Systems that do not use CRI checkpoint restore are not affected. This issue is fixed in versions 2.2.7 and 2.3.4.7d CVE-2026-415107.2 HIG

41.7%

——13OWASP Coraza WAF is a golang modsecurity compatible web application firewall library. From 3.0.0 until 3.8.1, Coraza silently drops arguments after SecArgumentsLimit in internal/corazawaf/transaction.go, while internal/bodyprocessors/urlencoded.go writes ARGS_POST without enforcing the cap. The flaw is triggered when an unauthenticated attacker sends more arguments than SecArgumentsLimit or supplies bodies that expand beyond the configured argument and flattening budgets. An unauthenticated attacker can flood requests with arguments so excess values are silently omitted from ARGS, ARGS_GET, ARGS_NAMES, ARGS_GET_NAMES, ARGS_POST, or ARGS_PATH before security rules inspect them; affected urlencoded and JSON processing can also consume excessive memory. The omission can bypass WAF rules protecting the downstream application, and unbounded or amplified argument processing can cause memory consumption and service degradation. This issue is fixed in version 3.8.1.7d CVE-2026-1095759.8 CRI

28.1%

——8In HAProxy Data Plane API 2.2.0 through 3.4.3 (and corresponding Enterprise Edition releases), Server.ConfigureAPI() has process-global authentication logic that improperly drops security controls across multiple transport types. When a TLS Certificate Authority (tls_ca) is configured and a numeric TLS port (tls_port) is specified, the application globally disables Basic authentication on its shared internal request handler. Because this handler is shared concurrently among HTTP, HTTPS, and UNIX domain socket listeners, and mutual TLS (mTLS) client certificate verification (tls.RequireAndVerifyClientCert) is enforced exclusively on the HTTPS server path, all other simultaneously active non-mTLS listeners fail open. Consequently, an unauthenticated attacker capable of reaching an affected HTTP or UNIX management endpoint can completely bypass credentials to gain full API administrative access. This allows unauthorized remote or local actors to read sensitive, cleartext HAProxy configuration data or replace the managed configuration entirely via the raw configuration endpoint. This has been fixed in versions 3.4.4, 3.3.9, 3.2.17, 3.0.25, 2.9.29, and their respective EE backports by migrating to a per-request TLS connection-state validation model.7d