Browser & mail CVEs.
Published CVEs that affect browser & mail software, ranked by exploit odds.
-
CVE-2026-125361
Tencent BrowserSkill through 0.4.2 contains a missing authentication vulnerability that allows local attackers to impersonate the bsk daemon because remoteSocket() connects to the loopback WebSocket without credentials or peer verification. Attackers who bind the daemon port before it starts or after it exits receive the extension's full tool surface, including evaluate, to run JavaScript through CDP in logged-in browser tabs.
-
CVE-2026-123976
EDFbrowser before 2.16 contains an out-of-bounds write vulnerability in check_edf_file() that allows attackers to overflow the 256-element annot_ch array by supplying excessive annotation signals. Attackers can craft an EDF+ or BDF+ file with more than 256 annotation signals to overwrite edfhdrblock members, including the signal parameter pointer, crashing the application when opened.
-
CVE-2026-106484
Insufficient policy enforcement in chrome-devtools-mcp in Google Chrome on Linux prior to 1.5.0 allowed a local attacker to perform arbitrary read/write via a malicious file. (Chromium security severity: Medium)
-
CVE-2026-131052
Race condition in UI in Google Chrome on on Android prior to 158.0.8127.36 allowed a local attacker leveraging social engineering to obtain cross-origin data via a co-installed app. (Chromium security severity: Medium)
-
CVE-2026-121109
Inadequate encryption strength in SSL in Google Chrome prior to 157.0.8103.38 allowed a remote attacker to potentially bypass web origin policy via crafted network traffic. (Chromium security severity: Low)
-
CVE-2026-118673
KeePassXC is a cross-platform community-driven port of the Windows application KeePass Password Safe. From 2.3.0 until 2.8.0, KeePassXC browser integration encrypts authorized responses in src/browser/BrowserMessageBuilder.cpp using a fixed 1 MiB output vector even though Botan::Sodium::crypto_box_easy() writes the plaintext length plus crypto_box_MACBYTES. A browser-protocol client that can access the per-user IPC endpoint, has a valid database association, and causes an unlocked database to generate and encrypt a sufficiently large authorized response can trigger a sequential out-of-bounds write before any response is delivered. This issue is fixed in version 2.8.0.
-
CVE-2026-76297
In Splunk Enterprise versions below 10.4.4, 10.2.8, 10.0.11, and 9.4.16, an unauthenticated user could trick a user that holds a role with the edit_federated_providers capability into turning off all federated search providers. Successful exploitation could temporarily prevent federated searches from retrieving data from remote datasets. The vulnerability is possible because the federated provider Representational State Transfer (REST) API accepts GET requests for actions that change provider state. The vulnerability requires the attacker to phish the affected user by tricking them into initiating a request within their browser. The unauthenticated user should not be able to exploit the vulnerability at will. For more information see Run federated searches over remote Splunk platform deployments (https://help.splunk.com/en/splunk-enterprise/search/federated-search/10.6/run-federated-searches-across-other-splunk-deployments/run-federated-searches-over-remote-splunk-platform-deployments) and Define roles on the Splunk platform with capabilities (https://help.splunk.com/en/splunk-enterprise/administer/manage-users-and-security/10.4/manage-splunk-platform-users-and-roles/define-roles-on-the-splunk-platform-with-capabilities) in the Splunk documentation.
-
CVE-2026-75518
DOMShell is a Chrome automation project whose MCP server exposes Unix-style tools for AI agents. Prior to 2.0.8, the DOMShell MCP server classifies the domshell_eval tool and eval command as read-tier operations that can run without --allow-write, even though eval shares the Runtime.evaluate dispatch path with the write-tier js command and has no side-effect gate. An MCP caller with tool-session access can therefore use domshell_eval to mutate DOM, window, or global state while the host approval interface, server authorization flags, denial messages, and audit records represent the operation as read-only. The bypass does not expose a capability beyond domshell_js for callers already authorized to write, but it violates the configured permission boundary for callers lacking that authorization. This issue is fixed in version 2.0.8.
-
CVE-2026-131089
Buffer overflow in CrashReporting in Google Chrome on on Windows prior to 158.0.8127.36 allowed a local attacker who had compromised the renderer process to execute arbitrary code outside the sandbox via a local program. (Chromium security severity: High)
-
CVE-2026-131016
Incorrect authorization in DevTools in Google Chrome prior to 158.0.8127.36 allowed a local attacker to bypass system access restrictions via a crafted HTML page. (Chromium security severity: Low)
-
CVE-2026-126862
Improper input validation in Core in Google Chrome on on Android prior to 157.0.8103.78 allowed a local attacker to bypass web origin policy via a co-installed app. (Chromium security severity: High)
-
CVE-2026-121049
Out of bounds read in V8 in Google Chrome prior to 157.0.8103.38 allowed a local attacker to read memory via a crafted file. (Chromium security severity: Medium)
-
CVE-2026-65587
A web page that contains unusual WebGPU content loaded into the GPU GLES render process can trigger an OOB write in the GPU user-space driver, leading to memory corruption and possible browser/GPU process crash. The software writes data past the end of an intended memory offset. Writing outside the bounds of allocated memory can corrupt adjacent memory, crash the process, or cause undefined behaviour.
-
CVE-2026-84786
Issue summary: A DANE-enabled TLS client may crash by dereferencing a NULL pointer when processing an unusual combination of certificate and TLSA records. Impact summary: The impact is limited to a client crash (potential Denial of Service). CWE: CWE-476: NULL Pointer Dereference Description: A DANE-enabled TLS client that supports TLSA records with at least one of the PKIX-TA(0) or PKIX-EE(1) usages alongside the DANE-TA(2) usage may crash by dereferencing a NULL pointer if the server claims a trusted CA's certificate to be its end-entity certificate and has an unusual mix of associated TLSA records. DANE SMTP clients should not be affected since only the DANE-TA(2) and DANE-EE(3) TLSA record usages are specified applicable in SMTP. For example, the Postfix MTA is not affected, while the Exim MTA does not at present use OpenSSL's DANE implementation. Use of DANE in validation of RFC7250 raw public keys (RPKs) is also unaffected, as these only use DANE-EE(3) TLSA records. Only applications that support combining the DANE PKI with a traditional PKI are potentially affected. A DANE-enabled browser or similar, that uses OpenSSL's DANE API could be affected. FIPS impact: no The DANE API is outside the FIPS module boundary.
-
CVE-2026-121199
Missing authorization in Chromoting in Google Chrome on on Mac prior to 157.0.8103.38 allowed a local attacker to obtain sensitive information via a local program. (Chromium security severity: Medium)
-
CVE-2026-121064
Improper input validation in Performance in Google Chrome prior to 157.0.8103.38 allowed a local attacker to potentially execute arbitrary code inside the sandbox via a local program. (Chromium security severity: Low)
-
CVE-2026-131032
Confused deputy in UI in Google Chrome on on Android prior to 158.0.8127.36 allowed a local attacker leveraging social engineering to obtain sensitive information via a co-installed app. (Chromium security severity: Medium)
-
CVE-2026-121232
Incorrect authorization in Core in Google Chrome prior to 157.0.8103.38 allowed a local attacker leveraging social engineering to obtain sensitive information via a co-installed app. (Chromium security severity: Low)
-
CVE-2026-131006
Race condition in CredentialProvider in Google Chrome on on Windows prior to 158.0.8127.36 allowed a local attacker to potentially execute arbitrary code outside the sandbox via a local program. (Chromium security severity: Medium)
-
CVE-2026-101178
Permission control vulnerability in the autofill service module. Impact: Successful exploitation of this vulnerability may affect service confidentiality.
-
CVE-2026-133652
Flowintel , contained a session management vulnerability in its browser authentication mechanism. The session identifier used a bare numeric user ID rather than a credential-bound token, resulting in several related weaknesses: - Remembered logins (remember-me cookies) remained valid after a user changed their password, allowing an attacker holding a previously issued cookie to retain access indefinitely. - Server-side session identifiers were not rotated upon login (local, SSO, or remembered), enabling session fixation. - Self-service email and password changes did not require verification of the current password, permitting an attacker with an active session to modify account credentials without knowledge of the existing password. - Administrator-initiated password resets did not revoke other browser sessions, invalidate remembered logins, or rotate the affected user's API key. An attacker who obtained a valid session or remember-me cookie (e.g., through XSS, network interception, or cookie theft) could maintain unauthorized access even after the legitimate user changed their password. Additionally, an attacker with any valid session could change the account's email address or password without knowing the current credentials, effectively taking over the account. Affected versions: Flowintel prior to 4.1.0.
-
CVE-2026-77314
Flyto2 Core is an execution kernel for automation and AI-agent workflows. Prior to 2.26.8, the MCP run_recipe tool in src/core/mcp_handler.py accepts a caller-controlled recipe_name that load_recipe() in src/cli/recipe.py appends to the src/recipes path without resolved-path containment. Parent-directory segments can therefore make Flyto load and execute a YAML workflow outside the bundled recipe directory, bypassing the boundary represented by list_recipes(). This does not necessarily provide unrestricted code execution because available workflow modules remain subject to Flyto's module policy, but it permits unintended execution of allowed browser, network, file, and data-processing modules. This issue is fixed in version 2.26.8.
-
CVE-2026-76543
In Splunk Enterprise versions below 10.6.1, 10.4.5, 10.2.9, 10.0.12, and 9.4.17, a user who holds the "power" Splunk role could store attacker-controlled Search Processing Language (SPL) in a shared data model. The vulnerability is possible because the Data Model Editor does not properly neutralize data model field names when it builds a preview search. The vulnerability requires the attacker to phish the affected user by tricking them into initiating a request within their browser. The attacker should not be able to exploit the vulnerability at will. For more information see Add a regular expression field (https://help.splunk.com/en/splunk-enterprise/manage-knowledge-objects/knowledge-management-manual/10.4/define-data-model-dataset-fields/add-a-regular-expression-field), Define roles on the Splunk platform with capabilities (https://help.splunk.com/en/splunk-enterprise/administer/manage-users-and-security/10.4/manage-splunk-platform-users-and-roles/define-roles-on-the-splunk-platform-with-capabilities), and About the search language (https://help.splunk.com/en/splunk-enterprise/search/search-manual/10.4/search-overview/about-the-search-language) in the Splunk documentation.
-
CVE-2026-131258
In Splunk SOAR versions below 8.8.0, a user with edit permission for a container or workbook task could store attacker-controlled Hypertext Markup Language (HTML) in a note. When another user with view permission for the affected container and its notes opens the affected note in Mission Control, the note could run unauthorized JavaScript code in the browser of that user, expose data available through the authenticated session, and perform actions available through the authenticated session. The vulnerability is possible because Splunk SOAR does not neutralize active content in all notes before Mission Control renders them.
-
CVE-2026-131260
In Splunk SOAR versions below 8.8.0, a user with the containers:edit permission could import a container that includes a malicious Hypertext Markup Language (HTML) note. When another user views the imported note, Splunk SOAR could run unauthorized JavaScript in the browser of that user, exposing data and allowing actions available to that user. The vulnerability is possible because Splunk SOAR does not sanitize HTML note content during container import before it renders the content in Mission Control.
-
CVE-2026-131234
In Splunk Enterprise Security versions below 8.7.0, an unauthenticated user could create and deliver an Entity Analysis dashboard link to a Splunk Enterprise Security user. If the Splunk Enterprise Security user opens the link, Splunk Enterprise Security could run attacker-controlled Search Processing Language (SPL) commands using the permissions of the Splunk Enterprise Security user. The commands could expose or modify some data available to the Splunk Enterprise Security user. The vulnerability is possible because the Entity Analysis dashboard inserts entity values from a link that an unauthenticated user creates into SPL searches without safely escaping the values. The vulnerability requires the attacker to phish the affected user by tricking them into initiating a request within their browser. The unauthenticated user should not be able to exploit the vulnerability at will. For more information see Entity analysis dashboard (https://help.splunk.com/en/splunk-enterprise-security-8/user-guide/8.7/exposure-analytics-and-ueba/entity-analysis-dashboard) in the Splunk documentation.
-
CVE-2026-131208
In Splunk Enterprise Security versions below 8.7.0, an unauthenticated user could create and deliver a crafted Investigation Workbench link to an authenticated Splunk Enterprise Security user. If the authenticated user opens the link, Investigation Workbench could run arbitrary Search Processing Language (SPL) commands using the permissions of the authenticated user. With this access, the unauthenticated user could expose or modify data available to the authenticated user. The vulnerability is possible because Investigation Workbench does not safely neutralize special characters in artifact values before adding the values to searches. The vulnerability requires the attacker to phish the affected user by tricking them into initiating a request within their browser. The unauthenticated user should not be able to exploit the vulnerability at will. For more information see Investigate a potential security incident on the investigation workbench in Splunk Enterprise Security (https://help.splunk.com/en/splunk-enterprise-security-7/user-guide/7.3/investigations/investigate-a-potential-security-incident-on-the-investigation-workbench-in-splunk-enterprise-security) and Users and roles for Splunk Enterprise Security (https://help.splunk.com/en/splunk-enterprise-security-8/install/8.7/installation/users-and-roles-for-splunk-enterprise-security) in the Splunk documentation.
-
CVE-2026-131201
In Splunk Enterprise Security versions below 8.7.0, an unauthenticated user with network-path interception capability could cause attacker-controlled JavaScript to run in the browser of a security analyst who previews a Portable Document Format (PDF) attachment in Mission Control. The JavaScript could expose data and perform actions available to the security analyst. The vulnerability is possible because Mission Control loads the PDF rendering worker from an external content delivery network. The vulnerability requires the unauthenticated user to intercept the network path to the external content delivery network and a security analyst to preview a PDF attachment in Mission Control. The unauthenticated user should not be able to exploit the vulnerability at will. For more information see Collaborate on investigations in Splunk Enterprise Security (https://help.splunk.com/en/splunk-enterprise-security-8/administer/8.7/investigations/collaborate-on-investigations-in-splunk-enterprise-security) in the Splunk documentation.
-
CVE-2026-131182
In Splunk Enterprise versions below 10.4.5, 10.2.9, 10.0.12, and 9.4.17, a user who holds the "power" Splunk role could store malicious Search Processing Language (SPL) in a shared dashboard. The malicious SPL could run using the permissions of the user who opened the dashboard and could access or modify data available to that user and disrupt Splunk Enterprise availability. The vulnerability is possible because Splunk Enterprise does not preserve quoted search argument boundaries when it processes a dashboard search. The vulnerability requires the attacker to phish the affected user by tricking them into initiating a request within their browser. The attacker, a user who holds the "power" Splunk role, should not be able to exploit the vulnerability at will.
-
CVE-2026-131184
In Splunk Enterprise versions below 10.6.1, 10.4.5, 10.2.9, 10.0.12, and 9.4.17, a low-privileged user who does not hold the "admin" or "power" Splunk roles could inject a malicious script into the name of a private saved search. If another user subsequently enables, disables, or edits the corresponding platform alert in the Monitoring Console, the injected script executes within that user's browser session, potentially allowing the low-privileged user to view sensitive information or modify data in Splunk Enterprise. The vulnerability is possible because the Monitoring Console does not safely handle saved-search names in the platform alert workflow. For more information see Enable and configure platform alerts (https://help.splunk.com/en/splunk-enterprise/administer/monitor/10.4/configure-the-monitoring-console/enable-and-configure-platform-alerts) and About configuring role-based user access (https://help.splunk.com/en/splunk-enterprise/administer/manage-users-and-security/10.4/manage-splunk-platform-users-and-roles/about-configuring-role-based-user-access) in the Splunk documentation.
-
CVE-2026-81434
A reflected Cross-Site Scripting (XSS) vulnerability in WatchGuard Dimension's Device Configuration History page allows an authenticated attacker with view permission on a device to execute arbitrary JavaScript in another authenticated user's browser by enticing the victim to open a specially crafted link.
-
CVE-2026-130158
RepoManager is a self-hosted deb/rpm package repository. Prior to 5.13.3, a registered host or a party with its host/API credential can submit package metadata. Inventory reporting accepts package name and version, while available-update reporting accepts package name, version, and repository; the repository field exists only on that path. RepoManager stores these values, and stored values are rendered on the host's package detail page without preventing stored HTML or JavaScript from executing. An administrator must later view the page for script execution. The script runs in the administrator's authenticated browser session and can expose administrative data or forge privileged requests; the advisory describes possible escalation through admin-only functionality. Fixed in 5.13.3.
-
CVE-2026-77359
Jupyter MCP Server is a Model Context Protocol server for Jupyter. Prior to 1.0.3, standalone streamable-http mode validates Host and Origin values for /mcp but does not apply the same browser-origin controls to the state-changing /api/connect and /api/stop management routes. A malicious website can use DNS rebinding so that its hostname resolves to a loopback, LAN, VPN, or published-container MCP listener, then send a browser-shaped request that /api/connect accepts to replace runtime_url and document_url values and start a kernel client against attacker-controlled infrastructure. The attacker can thereby cause Jupyter API and WebSocket traffic to reach that infrastructure and can use /api/stop to stop the current notebook kernel. The default broad bind can increase exposure, but the vulnerability is the missing management-route origin boundary, not a lack of Host validation on /mcp. This issue is fixed in version 1.0.3.
-
CVE-2026-129517
A cross-site scripting vulnerability exists in Win Bird JYUGYO SHIEN browser extensions for Google Chrome and Microsoft Edge, which allows an attacker to execute arbitrary script in the user's web browser.
-
CVE-2026-125346
Use after free in XML in Google Chrome prior to 157.0.8103.50 allowed a remote attacker to execute arbitrary code inside the sandbox via a crafted HTML page. (Chromium security severity: High)
-
CVE-2026-121101
Missing authorization in DevTools in Google Chrome prior to 157.0.8103.38 allowed a remote attacker leveraging social engineering to bypass system access restrictions into a privileged page via a crafted Chrome extension. (Chromium security severity: Low)
CVEs that affect browser & mail, sorted by how likely they are to be exploited (EPSS).