Roxy-WI is a web interface for managing Haproxy, Nginx, Apache and Keepalived servers. In versions 8.2.6.4 and prior, POST /config/versions/<service>/<server_ip>/<configver>/save interpolates the URL-path configver parameter directly into a config-version path that ends up at os.system(f"dos2unix -q {cfg}"). configver is not run through EscapedString (Pydantic doesn't validate path segments declared as str) and the surrounding .. block is the broken tuple-membership patch from GHSA-vapt-004. An authenticated user with role <= 3 ("user") therefore reaches a bin/sh -c command-injection sink. At time of publication, there are no publicly available patches.
Roxy-WI is a web interface for managing Haproxy, Nginx, Apache and Keepalived servers. In versions 8.2.6.4 and prior, GET /history/<service>/<server_ip> re-uses the server_ip path parameter as a user-id when service == 'user', with no authorization check. Any authenticated user — even a guest in an unrelated group — can list any other user's full action audit trail (server IPs touched, configs deployed, services restarted). At time of publication, there are no publicly available patches.
Roxy-WI is a web interface for managing Haproxy, Nginx, Apache and Keepalived servers. In versions 8.2.6.4 and prior, the /smon/agent/{version,uptime,status,checks}/<server_ip> family of routes takes the URL path component verbatim into requests.get(f'http://{server_ip}:{agent_port}/...'). The path component is constrained only by Flask's default URL converter, which permits any value (including IPv4 literals like 169.254.169.254, RFC1918 ranges, and 127.0.0.1). At time of publication, there are no publicly available patches.
Roxy-WI is a web interface for managing Haproxy, Nginx, Apache and Keepalived servers. In versions 8.2.6.4 and prior, wrap_line (app/modules/common/common.py:181-186) and highlight_word (app/modules/common/common.py:188-192) build raw HTML by string concatenation with no escaping. The frontend (app/static/js/script.js, log-viewer paths) uses .html(data) / .append(data) to inject the response body. Anyone able to write a line into a managed HAProxy/Nginx access log (i.e. anyone who can send an HTTP request to the public LB) can land an <svg/onload=…> payload that executes when a Roxy-WI admin opens the log viewer. At time of publication, there are no publicly available patches.
Roxy-WI is a web interface for managing Haproxy, Nginx, Apache and Keepalived servers. In versions 8.2.6.4 and prior, get_ldap_email (app/modules/roxywi/user.py:120-157) builds the LDAP search filter via f-string concatenation. The username URL path parameter is taken verbatim — no checkAjaxInput, no LDAP escape — and inserted, a username like *)(mail=*)(cn=* injects additional clauses, allowing the admin to enumerate or harvest attributes outside the intended record. At time of publication, there are no publicly available patches.
Roxy-WI is a web interface for managing Haproxy, Nginx, Apache and Keepalived servers. In versions 8.2.6.4 and prior, the HAProxy section-save endpoints (POST /api/service/haproxy/<server_id>/section/<section_type> and the PUT / global / defaults variants) accept a JSON option field that is not validated, not escaped, and is rendered verbatim into the generated HAProxy configuration via the section.j2, global.j2, and defaults.j2 Ansible templates. Because Roxy-WI then pushes the generated config to the load balancer and runs systemctl reload haproxy, an authenticated user with role ≤ 3 (user) can inject arbitrary HAProxy directives into the config that runs on every load balancer their group manages — including option external-check + external-check command /bin/bash -c '…', which gives remote code execution on the load balancer as the haproxy user on every health-check tick. At time of publication, there are no publicly available patches.
Roxy-WI is a web interface for managing Haproxy, Nginx, Apache and Keepalived servers. In versions 8.2.6.4 and prior, POST /waf/<service>/<server_ip>/rule/<rule_id>/save accepts a config_file_name form field that is passed straight through to config_mod.master_slave_upload_and_restart(...) as the destination path. The validation chain (_replace_config_path_to_correct → check_is_conf) only requires the path to contain a hard-coded service substring (nginx/haproxy/apache2/httpd/keepalived) and the substring conf or cfg, and to not contain ... The encoded-slash substitution 92 → / is applied before the substring check, so the attacker can build any absolute path anywhere on the LB filesystem as long as it satisfies those substring constraints. The body of the WAF rule (config form field) is written verbatim to that path. By choosing a filename like 92etc92cron.d92nginx_cfg_evil (resolving to /etc/cron.d/nginx_cfg_evil), an attacker drops a cron entry on the load balancer with attacker-controlled content. Cron parses the file on its next scan, executing the embedded job as root — full RCE on every load balancer the caller's group manages. At time of publication, there are no publicly available patches.
Roxy-WI is a web interface for managing Haproxy, Nginx, Apache and Keepalived servers. In versions 8.2.6.4 and prior, the install blueprint declares only bp.before_request → @jwt_required() (app/routes/install/routes.py:36-39). The individual endpoints install_exporter, install_waf, install_geoip, check_geoip, get_exporter_version, and get_task_status are not wrapped in page_for_admin and do not call roxywi_common.is_user_has_access_to_its_group(server_ip) or check_is_server_in_group(server_ip). Only the GET index page (install_monitoring) gates on roxywi_auth.page_for_admin(level=2). Because the missing decorators omit both role and group checks, any logged-in user — including the default guest role 4 — can install/reconfigure exporters, WAF, and GeoIP databases on every server in the Roxy-WI database, regardless of tenant ownership. The Ansible playbooks run with the per-server SSH credential stored in Roxy-WI, which the credentials' rightful owner (a different tenant) has provisioned with sudo rights for the management workflow. At time of publication, there are no publicly available patches.
Roxy-WI is a web interface for managing Haproxy, Nginx, Apache and Keepalived servers. In versions 8.2.6.4 and prior, PUT /smon/check (app/routes/smon/routes.py:117-138) gates only on roxywi_common.check_user_group_for_flask() — which validates that the caller has some group, not that the target check_id belongs to it. The downstream SQL update functions update_smon, update_smonHttp, update_smonTcp, update_smonPing, update_smonDns (app/modules/db/smon.py:515-562) all execute WHERE smon_id = ? with no user_group filter. The DELETE path is correctly filtered (app/modules/db/smon.py:319-327 does WHERE id = ? AND user_group = ?), demonstrating that the maintainers know the right pattern but did not apply it on UPDATE. Therefore any authenticated user can iterate over smon_id values and silently rewrite any other tenant's HTTP / TCP / Ping / DNS monitoring check. At time of publication, there are no publicly available patches.
Roxy-WI is a web interface for managing Haproxy, Nginx, Apache and Keepalived servers. In versions 8.2.6.4 and prior, agent_action (app/routes/smon/agent_routes.py:166-179) has decorators @bp.post('/agent/action/<action>') and @jwt_required() only — no role check, no group ownership check on the server_ip form field. Any authenticated user, including role 4 (guest), can start, stop, or restart the roxy-wi-smon-agent systemd unit on any server they can name. Roxy-WI executes the systemd action over its own SSH credentials (passwordless sudo), so the action runs as root on the target. At time of publication, there are no publicly available patches.
A heap buffer overflow flaw was found in 389 Directory Server. When serializing objectclass definitions, the oc_superior (SUP) field length is omitted from buffer size calculations in read_schema_dse() and schema_oc_to_string(), but the field is still written via strcat(). An attacker with Directory Manager privileges, or a compromised replication supplier, can trigger a server crash by creating objectclasses with long SUP values. This is an incomplete fix variant of CVE-2025-14905.
During an internal security assessment, a potential out-of-bounds write vulnerability was discovered in the BIOS of some ThinkPad products could allow a privileged local user to execute code in System Management Mode (SMM).
During an internal security assessment, a potential vulnerability was discovered in some ThinkPad embedded controller firmware that could allow a privileged local user to perform arbitrary reads or writes to privileged memory regions.
Jenkins 2.567 and earlier, LTS 2.555.2 and earlier does not encrypt secrets from POST config.xml submissions before storing them in job configurations unencrypted in job config.xml files on the Jenkins controller where they can be viewed by users with Item/Extended Read permission, or access to the Jenkins controller file system.
Jenkins 2.483 through 2.567 (both inclusive), LTS 2.492.1 through 2.555.2 (both inclusive) does not escape the user-provided description of a generic offline cause that could be set through the `POST config.xml` API, resulting in a stored cross-site scripting (XSS) vulnerability exploitable by attackers with Agent/Configure permission.
Jenkins 2.567 and earlier, LTS 2.555.2 and earlier does not ensure that the "from" parameter in the "Delegate to servlet container" security realm is safe to redirect to after login, allowing attackers to perform phishing attacks by redirecting users to an attacker-controlled domain.
Missing permission checks in Jenkins 2.567 and earlier, LTS 2.555.2 and earlier allow attackers with Overall/Read permission to determine other users' configured timezone and to enumerate view names of other users' "My Views".
A missing permission check in Jenkins 2.567 and earlier, LTS 2.555.2 and earlier allows attackers with Item/Cancel permission, but lacking Item/Read permission, to cancel queue items they do not have permission to view.
Jenkins 2.567 and earlier, LTS 2.555.2 and earlier improperly determines that a redirect URL after login is legitimately pointing to Jenkins when it contains tab or newline characters between `//`, allowing attackers to perform phishing attacks.
Jenkins 2.567 and earlier, LTS 2.555.2 and earlier improperly determines that a redirect URL after login is legitimately pointing to Jenkins when it contains relative path segments (`./` or `../`), allowing attackers to perform phishing attacks.
In Jenkins 2.567 and earlier, LTS 2.555.2 and earlier, it is possible for attackers to have Jenkins deserialize arbitrary types defined in Jenkins core or plugins from an attacker-controlled `config.xml` submission in a way that allows them to handle HTTP requests afterwards.
This can be used to impersonate any user and send HTTP requests on their behalf, up to and including use of the Script Console to run arbitrary code, or to read arbitrary files from the Jenkins controller.
Ghidra before 12.1.1 contains an uncontrolled memory allocation vulnerability in the Mach-O binary parser that allows attackers to cause denial of service. An attacker can supply a crafted Mach-O binary with an arbitrarily large ncmds load command count value, forcing the parser to allocate excessive heap memory without validating file size, crashing the Ghidra JVM.
Ghidra before 12.1 contains a SQL injection vulnerability in BSim filter types that concatenate user-supplied values directly into SQL queries without escaping or parameterization. Remote attackers can inject arbitrary SQL via the BSim network query protocol to read, modify, or delete data in the PostgreSQL database.
Ghidra before 12.1 contains a heap-use-after-free vulnerability in the decompiler's HighVariable::merge() function during the variable merging pass. Attackers can trigger this vulnerability by crafting a binary that causes stale pointers in the HighIntersectTest::highedgemap cache to be dereferenced, reading and writing the flags field of freed heap memory when a user opens the binary in Ghidra's decompiler view.
Ghidra before 12.2 contains an unauthenticated path traversal vulnerability in the IsfServer that accepts TCP connections and passes client-supplied namespace strings directly to filesystem operations without validation. Remote attackers can connect to port 54321 and send crafted protobuf messages with traversal sequences to enumerate filesystem paths and probe arbitrary files.
Ghidra before 12.0.4 contains a path traversal vulnerability in the theme import functionality that allows attackers to write files outside the intended theme directory. Attackers can craft malicious theme ZIP files with traversal sequences in filenames to execute arbitrary code or modify sensitive files like .bashrc or .ssh/authorized_keys.
Ghidra before 12.1 contains an authentication bypass vulnerability in PKIAuthenticationModule.authenticate() that allows any user with a valid CA-signed certificate to impersonate other users by presenting their public certificate with a null signature. Attackers can escalate privileges, modify repository access controls, exfiltrate shared reverse engineering databases, and permanently compromise server integrity.
Ghidra before 12.0.3 contains an out-of-memory vulnerability in the rust_demangle function that allocates unbounded output buffers without size limits. Attackers can craft malicious Rust symbol names in binaries to trigger exponential memory allocation, causing process crashes during binary analysis.
Ghidra before 12.0.2 contains a path traversal vulnerability in the extension installer that fails to validate ZIP entry names during extraction. Attackers can craft malicious extensions with traversal sequences like ../ in filenames to write arbitrary files outside the intended directory, enabling code execution.
Ghidra before 12.1 contains an unsafe deserialization vulnerability in client-side Shared-Project RMI connection code that allows unauthenticated remote code execution. Attackers can craft a malicious project file with a ghidra:// URL that, when opened via File → Open Project, deserializes untrusted objects using a Jython 2.7.4 gadget chain to execute arbitrary commands.
Ghidra before 12.1 contains a command injection vulnerability in URL annotation handling on Windows where cmd.exe metacharacters are not properly escaped. Attackers can execute arbitrary commands under the Ghidra user's privileges by embedding malicious URLs in program comments that victims click.
Ghidra 11.0 before 12.1 contains a SQL injection vulnerability in the changePassword() method of PostgresFunctionDatabase that fails to escape double quotes in usernames interpolated into ALTER ROLE statements. Authenticated attackers can inject SQL commands via crafted username parameters in PasswordChange network messages to escalate to PostgreSQL superuser privileges and gain full database control.
Ghidra before 12.1 contains a path traversal vulnerability in SameDirDebugInfoProvider that fails to validate filenames from ELF binary .gnu_debuglink sections before constructing file paths. Attackers can craft malicious ELF binaries with traversal sequences to probe filesystem existence and leak CRC32 hashes of arbitrary files during automatic DWARF analysis.
Ghidra before 12.1 contains a heap-use-after-free vulnerability in SleighBuilder::generatePointerAdd caused by iterator invalidation when PcodeCacher::allocateInstruction reallocates the issued vector. Attackers can trigger memory corruption by decompiling malicious binaries through the public Sleigh::oneInstruction C++ API, affecting downstream SLEIGH library consumers.
Ghidra 10.2 before 12.1 contains an uncontrolled resource consumption vulnerability in ExportTrie.parseTrie() that lacks cycle detection when traversing Mach-O binary export tries. A crafted Mach-O binary with circular references in the export trie causes unbounded queue growth and exponential string concatenation, triggering OutOfMemoryError that crashes the entire JVM and loses all unsaved work.
Improper Neutralization of Input During Web Page Generation ('Cross-site Scripting') vulnerability in WPZOOM Portfolio allows Reflected XSS.
This issue affects WPZOOM Portfolio: from n/a through 1.4.21.
image-size through 2.0.2 contains a denial of service vulnerability that allows remote attackers to permanently block the Node.js event loop by supplying a specially crafted ICNS image buffer. Attackers can craft an ICNS buffer containing valid magic bytes and a zero-valued entry length field to trigger an infinite loop in the ICNS parser, as the offset is never incremented when the entry length field is 0, causing the while loop condition to remain true indefinitely.
image-size through 2.0.2 contains a denial of service vulnerability that allows remote attackers to permanently block the Node.js event loop by supplying a specially crafted image buffer with a zero-valued size field in a recognized box-type. Attackers can trigger an infinite loop in the JXL or HEIF image parsers by providing a crafted image containing a box with a size of zero, causing the offset to never advance and permanently hanging the application.
Ghidra before 11.2 contains a use after free vulnerability in the Sleigh backend caused by undefined static initialization order of the SleighArchitecture::translators and XmlArchitectureCapability singletons. Attackers can trigger an infinite loop or denial of service during shutdown by exploiting the unsafe destruction order that causes iteration over deallocated memory.
Slate Digital Connect 1.37.0 for macOS installs a privileged helper tool, com.slatedigital.connect.privileged.helper.tool, which exposes the XPC service com.slatedigital.connect.privileged.helper.tool2. The helper validates connecting XPC clients by obtaining the client's process identifier and using it to retrieve code-signing information for the process. This PID-based client validation is subject to a time-of-check time-of-use race condition because process identifiers can be reused. A local attacker can exploit PID reuse so that validation is performed against a trusted process instead of the original connecting process. This allows unauthorized access to privileged helper functionality and may lead to local privilege escalation.
Slate Digital Connect 1.37.0 for macOS installs a privileged helper tool, com.slatedigital.connect.privileged.helper.tool, which exposes the XPC service com.slatedigital.connect.privileged.helper.tool2. The helper validates connecting XPC clients by checking only the subject.OU value of the client's signing certificate and does not verify that the certificate chains to a trusted code-signing authority. A local attacker can sign a malicious client with a self-signed certificate containing the expected organizational unit value and connect to the privileged XPC service. This allows unauthorized access to privileged helper functionality and may lead to local privilege escalation.
An HTML injection vulnerability in the "fetch links" email sent by Thinkst Applied Research Canarytokens, enabling Interface Manipulation, Cross-Site Scripting (XSS) in emails clients that render HTML emails.
This issue affects Canarytokens: from Docker tag sha-c0f3cf142 before sha-08c3f93d, from Git commit c0f3cf142 before 08c3f93d.
The Newsletters plugin for WordPress is vulnerable to time-based SQL Injection via the ‘wpmlsubscriber_id’ parameter in all versions up to, and including, 4.13 due to insufficient escaping on the user supplied parameter and lack of sufficient preparation on the existing SQL query. This makes it possible for unauthenticated attackers to append additional SQL queries into already existing queries that can be used to extract sensitive information from the database.
Debusine is an integrated solution to build, distribute and maintain a Debian-based distribution. Debian source packages (.dsc) and upload artifacts (.changes) are manifest files that name the files that make up the artifact. The parser used to read these files in Debusine accepted arbitrary fully user-controlled paths. The mergeuploads task could be abused to create arbitrary symbolic links on a worker, overwriting any file that the worker user has access to.
Debusine is an integrated solution to build, distribute and maintain a Debian-based distribution. Files managed by debusine are organized into artifacts. The endpoints that create and delete relationships between artifacts enforced no permissions checks beyond being able to see the artifacts in question.
The Doctreat Core plugin for WordPress is vulnerable to Privilege Escalation in all versions up to, and including, 1.6.8. This is due to the doctreat_process_registration() function not properly restricting the roles that a user can register with. This makes it possible for unauthenticated attackers to register as an administrator user.
The Easy Image Collage plugin for WordPress is vulnerable to Stored Cross-Site Scripting via 'grid[properties][borderColor]' and 'grid[images][N][attachment_url]' Parameters in all versions up to, and including, 1.13.6 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with author-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Because the data is stored via update_post_meta() rather than wp_insert_post() post content, WordPress's unfiltered_html restriction does not apply, meaning Authors cannot be blocked from this attack path by capability controls alone.
The MW WP Form plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'memo' parameter in all versions up to, and including, 5.1.3 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with editor-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. Because the memo value is stored via update_post_meta() rather than wp_insert_post(), WordPress's built-in kses and unfiltered_html protections do not apply, allowing attackers to break out of the textarea element via injected closing tags regardless of role-based content filtering.
The aThemes Addons for Elementor plugin for WordPress is vulnerable to Stored Cross-Site Scripting via 'title_tag' Widget Setting in all versions up to, and including, 1.1.8 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with contributor-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. This affects the Posts Timeline widget as well as the Posts Carousel widget across its default, Banner, and Modern skins, all of which omit the whitelist validation that is correctly applied in the Posts List widget.