Base de CVEs

Znuny through 7.3.6 constructs a ticket-search database query by directly concatenating user-supplied input into the SQL statement. This lack of proper sanitization or the use of parameterized queries allows an attacker to inject arbitrary SQL commands, leading to data theft, modification, or denial of service. Specifically this is in Kernel::System::Ticket::TicketSearch and Kernel::System::Ticket::ArticleSearchIndex::DB via the ContentSearch parameter. The parameter is expected to be either "AND" or "OR", but its value was concatenated directly into the generated SQL query without validation.

In DragonFly BSD through 6.4.2 before 4bb22ff, the ARP input handlers arpintr() and in_arpinput() (sys/netinet/if_ether.c) do not validate the hardware address length (ar_hln) and protocol address length (ar_pln) fields of incoming ARP packets. in_arpreply() then uses these attacker-controlled values as memcpy lengths when copying from kernel buffers that are only 6 and 4 bytes long, resulting in an out-of-bounds read of kernel heap and stack memory that is copied into the ARP reply sent back to the requester. An unauthenticated attacker on the adjacent network can use a single crafted ARP request to disclose up to roughly 250 bytes of each type of kernel memory, including kernel pointers. This is expected to be fixed in the 6.6 stable release when that is available.

A stack-based buffer overflow vulnerability in HPE NonStop TELSERV can cause memory corruption that may result in DoS.

Netgate pfSense FRR package (pfSense-pkg-frr) 2.0.2_1 and source revision 2.1.2_1 contain an OS command injection vulnerability in the FRR status filter. An authenticated WebGUI user with only the page-status-frr privilege can supply a filter value that breaks out of a single-quoted sed expression into which escapeshellarg() output is concatenated. Shell command substitution is then evaluated by a popen() pipeline, allowing attacker-selected commands to execute as root. There are updated FRR packages for pfSense Plus 26.07/26.03.1 and CE 2.9.0/2.8.1.

Backdrop CMS Core 1.35.x before 1.35.2 and 1.34.x before 1.34.6 mishandles tokens. A token is a small placeholder string (like [node:title] or [user:name]) used to automatically insert dynamic data into text. A browser is provided for selecting tokens by type. The product doesn't sufficiently limit access to the types of tokens that are available in the token browser. The names of the individual tokens are adequately protected.

MoviePilot through 3.1.5 contains a command injection vulnerability that allows attackers to execute arbitrary OS commands by embedding malicious instructions in untrusted content processed through tools like browse_webpage, search_web, and torrent metadata parsers, which then get passed to the unrestricted ExecuteCommandTool via asyncio.create_subprocess_shell. Attackers can craft content that instructs the model to invoke execute_command with confirm_dangerous set to true, bypassing the validate_command_safety denylist and exploiting the admin session's is_admin context to achieve remote code execution on the host.

A vulnerability in HPE NonStop QIOMON allows an out-of-bounds write that can corrupt buffer data.

A vulnerability in HPE NonStop CIPSAM allows an out-of-bounds write that can corrupt buffer data.

A vulnerability in the finger daemon of HPE NonStop TCP/IP UTILITIES could allow an unauthenticated remote attacker to cause a denial of service.

IntentKit through 2.6.3 contains an authorization bypass vulnerability where the executor validates tool calls against the full tool registry rather than the per-request narrowed tool advertisement enforced by ToolBindingMiddleware, allowing owner-only state-changing tools such as wallet transfer, posting, memory-write, and call-agent to execute without any human-in-the-loop or approval gate. An unauthenticated or public caller in WEB, API, TWITTER, or TELEGRAM contexts can exploit this by supplying the exact de-advertised tool name and arguments via indirect prompt injection or by replaying checkpoint and history data on a reused thread, achieving unauthorized execution of owner-restricted capabilities.

A security vulnerability in HPE NonStop SNMP Trap Multiplexer may cause memory corruption

Ellucian WebAdvisor 3.2 contains an unauthenticated XML External Entity (XXE) injection vulnerability that allows remote attackers to submit crafted XML POST requests to the /WebAdvisor/WebAdvisor endpoint without disabling external entity and DTD resolution. Attackers can exploit this flaw to force the server into SSRF requests, coerce NTLMv2 hash disclosure via attacker-controlled SMB shares, and read arbitrary files from the underlying Windows host through an out-of-band error-based channel. WebAdvisor entered end-of-support status in June 2022 and users are encouraged to migrate to Colleague Self-Service.

SODOLA SL902-SWTGW124AS firmware before 200.2.12 contains a buffer overflow vulnerability that allows unauthenticated remote attackers to cause a denial of service by sending an oversized boundary value in the Content-Type header of a crafted POST request. Attackers can exploit the embedded HTTP server's multipart/form-data boundary parser to cause the device to become fully unresponsive — dropping all management access and traffic forwarding during its self-recovery cycle.

Heym from 0.0.55 before 0.0.121 contains a privilege escalation vulnerability that allows shared collaborators to overwrite a workflow's error_workflow_id because the field is missing from OWNER_ONLY_EXECUTION_FIELDS. Attackers can point the error workflow to their own workflow with an HTTP node, and when the parent workflow errors the owner's decrypted credentials are injected and exfiltrated.

Nodemailer 3.0.0 before 10.0.13 has a regular expression denial of service in SMTPConnection._actionEHLO, whose four AUTH mechanism detection regexes contain overlapping whitespace sub-patterns that backtrack quadratically. A malicious or man-in-the-middle SMTP server can reply to EHLO with 250-AUTH, a long whitespace run and a non-matching token, blocking the Node.js event loop for tens of seconds to hours before authentication.

Renovate before 44.79.0 contains a server-side request forgery vulnerability that allows users to make the shared HTTP client request arbitrary internal URLs via extends presets or customDatasources. Attackers can point a preset at an instance metadata service endpoint to obtain AWS credentials or leak sensitive information through Renovate logs.

Renovate before 44.60.1 contains an allowlist bypass vulnerability that allows repositories to set restricted environment variables by placing env inside a shared preset. Attackers controlling an externally sourced preset can override variables like PATH, LD_PRELOAD, NODE_OPTIONS, or GIT_SSL_CAINFO in child processes on self-hosted instances.

Renovate before 44.60.1 contains an allowlist bypass vulnerability that allows repository users to evade the allowedHeaders restriction by placing hostRules headers inside shared presets. Attackers can supply internal or externally sourced presets with unvalidated hostRules[].headers to inject custom HTTP headers into requests Renovate sends to other hosts.

Better Auth from 1.5.0-beta.12 before 1.7.7 contains an authentication bypass vulnerability in the OAuth Proxy plugin that allows unauthenticated attackers to sign in as other users by forging encrypted profiles. When the proxy shares the global secret and cookie-stored state is used, attackers can obtain encrypted state for a fake profile with a victim's verified email to take over the account via implicit linking.

Better Auth @better-auth/sso from 1.7.0-rc.2 before 1.7.3 contains an authentication bypass vulnerability that allows authenticated users to sign in as other users by registering a personal SSO provider. Attackers can copy a legitimate provider's OIDC issuer or SAML entity ID and assert a victim's subject or NameID, because account lookup ignores providerId, creating sessions for victim accounts.

GitPython before 3.2.0 contains an argument injection vulnerability in Head.checkout() that allows remote attackers to read arbitrary files by supplying a branch name parsed as a git option. Attackers can host a repository whose default branch is named --pathspec-from-file=<path>, causing cloning applications that check out the active branch to leak file contents through GitCommandError messages.

GitPython from 3.1.30 before 3.2.0 contains a remote code execution vulnerability in Remote.pull() whose unsafe-option guard inspects only keyword arguments, not the positional refspec. Attackers controlling the refspec can supply --upload-pack=<command> to execute OS commands as the calling process over local-path or SSH transports.

GitPython before 3.2.0 contains a regular expression denial of service vulnerability in the _re_actor_epoch pattern used by parse_actor_and_date on author, committer, and tagger lines. Attackers can serve repositories with commit or tag objects containing long space-filled lines lacking a valid epoch or timezone, causing quadratic backtracking and CPU exhaustion.

GitPython before 3.2.0 contains a remote code execution vulnerability that allows attackers to run commands because Git.ls_remote forwards URLs without check_unsafe_protocols validation. Attackers can supply ext:: remote-helper URLs to execute commands when git permits the ext transport, while merge_base and IndexFile.move also accept unguarded option injection.

Parse Server before 8.6.97 and 9.0.0 before 9.10.2-alpha.6 does not handle errors raised while sending verification or password reset emails, causing the server process to exit. A client holding only the public application keys can sign up, request a verification email, or request a password reset while the mail provider rejects, rate limits, or is unavailable, crashing the server.