CVE Database

Pydantic AI is a Python agent framework for building Generative AI applications. In versions 1.65.0 through 1.105.0, and 2.0.0b1 through 2.0.0b5, a client that submits message history to a Pydantic AI UI adapter (such as the Vercel AI adapter) can reference arbitrary files in the application's model-provider or cloud-storage account. While file URL parts are validated against a scheme allowlist, UploadedFile references — which point to a file by provider file ID or cloud-storage URI (e.g. s3://…, gs://…) — were forwarded without validation. Because the provider resolves an UploadedFile using the server-side identity (IAM role, service account, or provider API key) rather than the client's, an attacker can craft message history to make the server read objects from its own account or other tenants, given a referenceable identifier. Exploitation requires a valid file identifier, which is not always unguessable depending on how the application names objects. This issue has been fixed in versions 1.106.0 and 2.0.0b6.

Pydantic AI is a Python agent framework for building Generative AI applications. In versions 1.56.0 through 1.98.0, when an application opts a URL into force_download='allow-local' (disabling the default block on private/internal IPs), the cloud-metadata blocklist could be bypassed by encoding the metadata IP in an IPv6 transition form (IPv4-mapped IPv6, 6to4, or NAT64), exposing cloud IAM short-term credentials on dual-stack or translated networks. This is an incomplete fix of GHSA-2jrp-274c-jhv3 / CVE-2026-25580, whose remediation did not hold for IPv6-encoded forms of the metadata IPs. An application is affected only if it explicitly opts a FileUrl (ImageUrl, AudioUrl, VideoUrl, DocumentUrl) into force_download='allow-local' on a URL influenced by untrusted input; it is not affected when using bundled integrations to ingest user input (Agent.to_web / clai web, VercelAIAdapter, AGUIAdapter / Agent.to_ag_ui), since they do not propagate force_download from external data, nor when downloading only from developer-controlled URLs. This issue has been fixed in version 1.99.0.

undici's retry interceptor can deliver a response whose body length does not match the Content-Length header exposed to the application after a retry or resume of a partial response. In undici before 6.28.0, from 7.0.0 up to before 7.29.0, and from 8.0.0 up to before 8.9.0, a malicious or faulty upstream can return a partial response with a mismatched framing header, close the socket early, and have the retry interceptor assemble a body of a different length while the original Content-Length stays attached. Applications that use the retry interceptor and forward upstream headers and bodies downstream, such as proxies or gateways, may then emit an invalid HTTP response with a stale Content-Length, leading to downstream response desynchronization, connection hangs, or response corruption. Exploitation requires the retry interceptor enabled, an upstream returning a mismatched partial response, and a downstream forwarder that does not remove or recalculate Content-Length. The issue is fixed in undici 6.28.0, 7.29.0, and 8.9.0.

Autel MaxiCharger AC Elite Home NFC Stack-based Buffer Overflow Arbitrary Code Execution Vulnerability. This vulnerability allows physically present attackers to execute arbitrary code on affected installations of Autel MaxiCharger AC Elite Home EV chargers. Authentication is not required to exploit this vulnerability. The specific flaw exists within the handling of card responses via the NFC interface. A crafted card response can trigger an overflow of a fixed-length stack-based buffer. An attacker can leverage this vulnerability to execute code in the context of the device. Was ZDI-CAN-29044.

Autel MaxiCharger AC Elite Home USB Heap-based Buffer Overflow Arbitrary Code Execution Vulnerability. This vulnerability allows physically present attackers to execute arbitrary code on affected installations of Autel MaxiCharger AC Elite Home EV chargers. Authentication is not required to exploit this vulnerability. The specific flaw exists within the handling of custom USB packets. The issue results from the lack of proper validation of the length of user-supplied data prior to copying it to a fixed-length, heap-based buffer. An attacker can leverage this vulnerability to execute code in the context of the device. Was ZDI-CAN-29048.

Autel MaxiCharger AC Elite Home USB Authentication Bypass Vulnerability. This vulnerability allows physically present attackers to bypass authentication on affected installations of Autel MaxiCharger AC Elite Home EV chargers. Authentication is not required to exploit this vulnerability. The specific flaw exists within the exposed USB interface. The issue results from the lack of authentication prior to allowing access to functionality. An attacker can leverage this vulnerability to bypass authentication on the system. Was ZDI-CAN-29046.

Autel MaxiCharger AC Elite Home Software Update Improper Verification of Cryptographic Signature Arbitrary Code Execution Vulnerability. This vulnerability allows physically present attackers to execute arbitrary code on affected installations of Autel MaxiCharger AC Elite Home EV chargers. Authentication is not required to exploit this vulnerability. The specific flaw exists within the handling of software updates. The issue results from the lack of proper validation of a user-supplied software update image. An attacker can leverage this vulnerability to execute code in the context of the device. Was ZDI-CAN-29062.

Sourcecodester Fantastic Blog CMS 1.0 is vulnerable to Cross Site Scripting (XSS) in pageEditMember.php via the address field.

GitLab has remediated an issue in GitLab CE/EE affecting all versions from 16.6 before 19.0.5, 19.1 before 19.1.3, and 19.2 before 19.2.1 that under certain conditions could have allowed an unauthorized user to view project import source information due to a missing authorization check.

linuxfabrik-lib provides Python modules for database access, caching, shell execution, and API integrations. Prior to version 6.0.0, lib.url.fetch() followed cross-origin redirects while forwarding caller-supplied credential headers other than Authorization and Cookie, allowing a malicious redirect-capable server to receive headers such as X-Auth-Token from authenticated monitoring requests. This issue is fixed in version 6.0.0.

Linuxfabrik monitoring-plugins provides Python monitoring plugins for Icinga, Nagios, and related monitoring systems. In version 6.0.0, the logfile check legacy database migration moved a predictable path from /tmp with os.rename() and allowed a local user controlling the plugin account to place a symlink that would be followed by sqlite3.connect() during a root-run check.

MCP Ruby SDK is the official Ruby SDK for Model Context Protocol servers and clients. Prior to 0.23.0, MCP::Server::Transports::StreamableHTTPTransport in the mcp gem does not expire sessions by default, so repeated initialize requests retain unbounded ServerSession objects and can exhaust process memory. This issue is fixed in version 0.23.0.

MCP Ruby SDK is the official Ruby SDK for Model Context Protocol servers and clients. Prior to 0.23.0, MCP::Server::Transports::StdioTransport and MCP::Client::Stdio in the mcp gem use IO#gets without a byte limit, allowing a peer that sends data without a newline to exhaust process memory. This issue is fixed in version 0.23.0.

MCP Ruby SDK is the official Ruby SDK for Model Context Protocol servers and clients. Prior to 0.23.0, MCP::Server::Transports::StreamableHTTPTransport in the mcp gem does not validate the HTTP Host or Origin request headers, which allows a malicious browser page to use DNS rebinding to reach a locally running MCP server and invoke exposed tools. This issue is fixed in version 0.23.0.

DriveLock Directory Traversal Information Disclosure Vulnerability. This vulnerability allows remote attackers to disclose sensitive information on affected installations of DriveLock. Authentication is required to exploit this vulnerability. The specific flaw exists within the web service, which listens on TCP port 4568 by default. The issue results from the lack of proper validation of a user-supplied path prior to using it in file operations. An attacker can leverage this vulnerability to disclose information in the context of the service account. Was ZDI-CAN-28713.

DriveLock Directory Traversal Information Disclosure Vulnerability. This vulnerability allows remote attackers to disclose sensitive information on affected installations of DriveLock. Authentication is not required to exploit this vulnerability. The specific flaw exists within the web service, which listens on TCP port 4568 by default. The issue results from the lack of proper validation of a user-supplied path prior to using it in file operations. An attacker can leverage this vulnerability to disclose information in the context of the service account. Was ZDI-CAN-28719.

GitLab has remediated an issue in GitLab CE/EE affecting all versions from 18.4 before 19.0.5, 19.1 before 19.1.3, and 19.2 before 19.2.1 that under certain conditions could have allowed an authenticated user with guest-role permissions to access test report contents they were not authorized to view due to improper access control enforcement.

GitLab has remediated an issue in GitLab CE/EE affecting all versions from 14.0 before 19.0.5, 19.1 before 19.1.3, and 19.2 before 19.2.1 that under certain conditions could have allowed an attacker to execute arbitrary JavaScript in another user's browser via a crafted URL, due to improper sanitization of user-controlled input.

Dify AI Workflow oauth_redirect_url Open Redirect Vulnerability. This vulnerability allows remote attackers to disclose sensitive information on affected installations of Dify. User interaction is required to exploit this vulnerability in that the target must visit a malicious page or open a malicious file. The specific flaw exists within the OAuth flow handling in the AppInitializer component. An attacker can force a redirection to a site that serves malicious content. An attacker can leverage this vulnerability to disclose information in the context of the application. Was ZDI-CAN-29196.

GitLab has remediated an issue in GitLab EE affecting all versions from 18.8 before 19.0.5, 19.1 before 19.1.3, and 19.2 before 19.2.1 that under certain conditions could have allowed some sensitive information to be disclosed to an unintended host due to improper handling of upstream requests in virtual registries.

GitLab has remediated an issue in GitLab EE affecting all versions from 19.1 before 19.1.3 and 19.2 before 19.2.1 that under certain conditions could have allowed an authenticated user to bypass administrator-configured tool governance policies due to improper authorization enforcement during token generation.

GitLab has remediated an issue in GitLab EE affecting all versions from 19.1 before 19.1.3 and 19.2 before 19.2.1 that under certain conditions could have allowed an authenticated user to access information from unauthorized projects due to improper neutralization of untrusted content processed by the AI-assisted code review functionality.

GitLab has remediated an issue in GitLab CE/EE affecting all versions from 8.8 before 19.0.5, 19.1 before 19.1.3, and 19.2 before 19.2.1 that under certain conditions could have allowed an unauthenticated user to view the title of a confidential issue through a publicly accessible merge request due to improper authorization checks.

GitLab has remediated an issue in GitLab CE/EE affecting all versions from 12.8 before 19.0.5, 19.1 before 19.1.3, and 19.2 before 19.2.1 that under certain conditions could have allowed an authenticated user with Maintainer role to modify protected branch configuration due to improper authorization in a projects API endpoint.

GitLab has remediated an issue in GitLab EE affecting all versions from 17.0 before 19.0.5, 19.1 before 19.1.3, and 19.2 before 19.2.1 that under certain conditions could have allowed an authenticated user to merge code into a protected branch without the required approvals due to a race condition in approval rule processing.

pip would incorrectly handle doubly-encoded package URLs from indexes allowing for files to be installed to arbitrary locations on disk even when installing wheels. This vulnerability requires downloading or installing a package from a malicious package index to succeed, malicious packages alone are not able to exploit this vulnerability. Note that this vulnerability only materially impacts users running `pip download` with the `--only-binary` option as installing source distributions from an untrusted index is already an unsafe operation that executes code during install time.

Netty is an asynchronous, event-driven network application framework. In versions prior to 4.1.136.Final and 4.2.16.Final, Netty's STOMP encoder ( StompSubframeEncoder ) does not escape or validate header values in  CONNECT  and  CONNECTED  frames, so raw newline ( \n ) characters in a header value are written directly to the wire, allowing an attacker who controls a header value to inject additional STOMP headers. This happens because the encoder intentionally skips escaping for CONNECT/CONNECTED frames per the STOMP 1.2 specification but never rejects the raw newlines, and since a broker parses each line as a separate header, an attacker controlling a value such as a user-supplied login or passcode can overwrite connection parameters or add authentication/role headers to bypass authentication or escalate privileges (the actual impact is broker-dependent). The issue is fixed in versions 4.1.136.Final and 4.2.16.Final.

Netty is an asynchronous, event-driven network application framework. In versions prior to 4.1.136.Final and 4.2.16.Final, Netty's HAProxy encoder ( HAProxyMessageEncoder ) writes AF_UNIX source and destination socket addresses into the HAProxy V1 text protocol without validating them for CRLF characters, so an attacker who controls an AF_UNIX address can inject  \r\n  sequences and split the single PROXY header into multiple lines. This is possible because the V1 protocol uses CRLF as its line terminator and, unlike IPv4/IPv6 addresses whose format checks implicitly reject CRLF, AF_UNIX addresses are only validated for length (up to 108 bytes), allowing a forged second PROXY header line that spoofs the client source/destination IP to a downstream server or load balancer. The issue is fixed in versions 4.1.136.Final and 4.2.16.Final.

Netty is an asynchronous, event-driven network application framework. Prior to versions 4.1.136.Final and 4.2.16.Final, Netty's HTTP/2-to-HTTP/1.x translation layer (`Http2StreamFrameToHttpObjectCodec` and `InboundHttp2ToHttpAdapter`) fails to deduplicate or validate `Host` headers when an HTTP/2 client supplies both the `:authority` pseudo-header and a literal `host` header in a single HEADERS frame. The translator maps `:authority` to `Host` and separately copies the literal `host` header, producing an `HttpRequest` object containing two `Host` headers with attacker-controlled differing values. This issue has been fixed in versions 4.1.136.Final and 4.2.16.Final.

MathLive provides web components for math display and input. Prior to 0.110.0, MathLive fails to escape text-mode content in \text{} and \mbox{} commands in Box.toMarkup at src/core/box.ts, in xmlEscape, scanText, and text-mode output in src/formats/atom-to-math-ml.ts, and through convertLatexToMarkup, convertLatexToMathMl, <math-span>, <math-div>, and the default identity MathfieldElement.createHTML, allowing malicious input to run arbitrary JavaScript when rendered. This issue is fixed in version 0.110.0.