CVE-20XX-?
agent?

Library CVEs.

Published CVEs that affect library software, ranked by exploit odds.

987 CVEs · updated 2026-11-19 02:05 UTC

Reset all
  1. CVE-2026-110747

    ProtobufParser._readStringKey() reads the length prefix of a protobuf map entry's string key into an attacker-controlled int and then checks (_inputPtr + len) <= _inputEnd using signed arithmetic. For values near Integer.MAX_VALUE the addition overflows to a negative value and the check passes even when the input contains no key data. The parser then calls _finishShortText(len), which expands the text buffer to the declared length via TextBuffer.expandCurrentSegment and Arrays.copyOf, throwing OutOfMemoryError: Requested array size exceeds VM limit. An 8-byte malformed message containing a map entry with a declared key length of 0x7FFFFFFF is sufficient. Configured StreamReadConstraints, including maxStringLength, maxNameLength, maxDocumentLength and maxNestingDepth, do not prevent the failure. Applications running with -XX:+ExitOnOutOfMemoryError terminate the JVM. This is the same integer-overflow pattern addressed for other call sites in the remediation associated with CVE-2026-104016 and CVE-2026-104017; the map-key call site was corrected on the 3.x branches only, leaving the corresponding 2.x branches unguarded. Release lines 2.18, 2.19 and 2.20 do not contain _readStringKey() and are not affected by this vector. No privileges or user interaction beyond the ability to submit data for parsing are required. The 3.0 release line and 3.1.0-3.1.5 / 3.2.0-3.2.1 do not contain _readStringKey() and are therefore not affected; the method first appears in 3.1.6 and 3.2.2, mirroring its introduction at 2.21.6 on the 2.x side.

    Memory safety Library

    7.5 HIGH 0.4% exploit odds Fix listed
  2. CVE-2026-114366

    An information disclosure vulnerability in the Qt Framework QWebSocket authentication handling of the Qt WebSockets module.

    Disclosure Library

    2.3 LOW 0.4% exploit odds No fix listed
  3. CVE-2026-61845

    Crass is a Ruby CSS parser that's fully compliant with the CSS Syntax Level 3 specification. Prior to 1.0.7, Crass converts `CSS scientific notation` number values with unbounded exponentiation before clamping the result to `Float::MAX`. Applications that use Crass to parse attacker-controlled CSS strings can be forced to spend disproportionate CPU and memory parsing a tiny input, potentially resulting in a crash. This issue is fixed in version 1.0.7.

    Denial of service Library

    8.9 HIGH 0.4% exploit odds Fix listed
  4. CVE-2026-97039

    Building a malicious package that uses "#cgo pkg-config:" directives could load untrusted pkg-config ".pc" files or inject unvalidated compiler and linker flags. The pkg-config flag allowlist previously permitted "--with-path=" and "--personality=", allowed "--define-variable=" values beginning with "=" or "$" (which can rewrite paths in system ".pc" files into sysroot-prefixed "-L=" or "-L$SYSROOT" linker search paths), and allowed package arguments ending in ".pc" or containing path separators or "..", which caused pkg-config to load local ".pc" files directly from the package directory. In addition, flags returned by "pkg-config --cflags" in cgoCPPFLAGS were not checked when determining whether internal linking is safe. The go command now disallows "--with-path=", "--personality=", leading "=" or "$" characters in "--define-variable=" values, and local ".pc" paths in "#cgo pkg-config:" directives, and inspects cgoCPPFLAGS when checking internal linking compatibility.

    Other Library

    8.3 HIGH 0.4% exploit odds No fix listed
  5. CVE-2026-117501

    HTTP::Body versions before 1.24_01 for Perl allow memory exhaustion via unbounded buffering of chunked request bodies. In chunked mode, used when no content length is given, add() holds each chunk in memory until all of it has arrived. A line that is not a chunk-size line, a size line without CRLF, or a size too large to fit an integer leaves the parser waiting forever, and every later add() grows the buffer. A remote attacker can exhaust memory in applications that pass chunked request bodies to HTTP::Body.

    Denial of service Library

    7.5 HIGH 0.4% exploit odds Fix listed
  6. CVE-2026-114620

    Authlib is a Python library which builds OAuth and OpenID Connect servers. Prior to 1.9.0, Authlib's RFC 9101 JWTAuthenticationRequest passes the unauthenticated, client-controlled request_uri parameter to the abstract get_request_object() hook before applying a destination or registration policy; the hook itself performs no network request, while the documentation demonstrates an unrestricted HTTP fetch. In deployments whose integrator implements that hook as an unrestricted outbound fetch, a remote attacker who knows a valid public client identifier can make the authorization server request internal, link-local, or otherwise unintended URLs and can distinguish fetch outcomes through different error responses and timing. Deployments whose custom hook already applies a secure allowlist, destination controls, and retrieval limits are not affected. The JWTAuthenticationRequest implementation in authlib/oauth2/rfc9101/authorization_server.py forwards request_uri to get_request_object() before an application-approved retrieval policy accepts the destination. This issue is fixed in version 1.9.0.

    Disclosure Library Authlib

    5.4 MEDIUM 0.4% exploit odds Fix listed
  7. CVE-2026-123019

    A weakness has been identified in eosphoros-ai DB-GPT up to 0.8.2. Impacted is the function exec_module of the file hf_adapter.py of the component HuggingFace Adapter. Executing a manipulation of the argument trust_remote_code can lead to inclusion of functionality from untrusted control sphere. The attack can be launched remotely. The exploit has been made available to the public and could be used for attacks. The project was informed of the problem early through an issue report but has not responded yet.

    Code execution Library

    5.5 MEDIUM 0.4% exploit odds No fix listed
  8. CVE-2026-96644

    Pillow is a Python imaging library. Prior to 13.0.0 on Windows, src/PIL/WmfImagePlugin.py passes unchecked 32-bit WMF or EMF dimensions to the native drawwmf handler in src/display.c. The handler sizes a DIB allocation from 16-bit-truncated bcWidth and bcHeight fields but copies pixel data using the original full dimensions. A crafted wide, short metafile can pass the pixel-count decompression-bomb check and make the Windows-only renderer read beyond the heap buffer, potentially disclosing memory or crashing the process. This issue is fixed in version 13.0.0.

    Memory safety Library Python Pillow

    7.1 HIGH 0.4% exploit odds Fix listed
  9. CVE-2026-114109

    uv before 0.11.15 contains a path traversal vulnerability that allows malicious wheels to write files outside the scripts directory via crafted console_scripts or gui_scripts entry point names. Attackers who induce users to install a malicious wheel can place executables in PATH directories, shadowing or overwriting existing binaries for code execution.

    Disclosure Library

    8.7 HIGH 0.4% exploit odds Fix listed
  10. CVE-2026-97054

    For HTTP/1 servers, when MaxBytesReader's limit is exceeded, "Connection: close" header is automatically set without synchronization, which might race with another header being set explicitly in the server handler. As a result, an HTTP/1 server that utilizes MaxBytesReader or MaxBytesHandler can crash due to runtime panic if its handler reads the request body concurrently with writing the response header.

    Denial of service Library

    5.9 MEDIUM 0.4% exploit odds No fix listed
  11. CVE-2026-126062

    Nokogiri is an open source XML and HTML library for the Ruby programming language. Prior to 1.19.5, on JRuby, applications that parse untrusted XML with DTDLOAD, DTDVALID, or NOENT can have the NONET restriction bypassed when an external DTD or entity uses a short-form IPv4 host or uses absolute paths resolved against a network URL. The resolver can treat these identifiers as local and send a network request, potentially enabling server-side request forgery. CRuby and default parse options are not affected. This issue is fixed in version 1.19.5.

    Disclosure Library

    3.7 LOW 0.4% exploit odds Fix listed
  12. CVE-2026-106554

    One CBOR-parsing issue in jackson-dataformats-binary's CBOR module: an unbounded run of CBOR tags drives quadratic-time TagList growth (CPU/GC-exhaustion DoS, PoC-proven). PoC executed against com.fasterxml.jackson.dataformat:jackson-dataformat-cbor:2.18.2; present in current 2.x and 3.x (tools.jackson.dataformat.cbor) HEAD. Distinct from the Avro and Smile/Protobuf advisories for this repo. Quadratic CPU/GC DoS from an unbounded chain of CBOR tags (PoC-proven). File: cbor/src/main/java/.../cbor/CBORParser.java (tag handling -> TagList). CBOR permits a data item to be preceded by any number of semantic tags. The CBOR parser accumulates them in a TagList whose backing array grows by a fixed increment (+8), not by doubling, and re-scans/copies on each growth. An attacker sends a value preceded by a long run of tag bytes; the fixed-increment growth makes accumulating N tags O(N^2) in time (plus GC pressure), with no cap on the number of stacked tags. PoC (executed, JDK 17): feeding a CBOR stream consisting of a long run of tag bytes (e.g. tag 0xC0+) before a trivial value forces quadratic CPU/GC; parse time grows super-linearly with the tag count, pinning a request thread with no configurable limit. Impact. Algorithmic-complexity denial of service (CWE-407): a modest input of stacked tags consumes quadratic CPU/GC, pinning a request thread; concurrent requests exhaust the pool. Fix. Grow TagList by doubling and/or cap the number of stacked tags (StreamReadConstraints-style).

    Denial of service Library

    7.5 HIGH 0.4% exploit odds Fix listed
  13. CVE-2026-104895

    The asynchronous Smile parser (NonBlockingByteArrayParser) uses a raw-binary length prefix taken from untrusted input to allocate a byte[] before the body is read, with no upper bound and no overflow check. On the raw-binary marker 0xFD, _decodeVInt() computes (value << 6) + (i & 0x3F) without the (value >>> 25) != 0 overflow guard present in the blocking SmileParser._readUnsignedVInt, and _startRawBinary then allocates the declared length immediately. A Smile feed of roughly ten bytes can therefore force an eager allocation of approximately 2 GB, producing an OutOfMemoryError on reactive or asynchronous Smile endpoints. A VInt crafted to overflow int yields a negative length and a NegativeArraySizeException, a RuntimeException that is not a JacksonException and so escapes the parser contract. Either outcome crashes the request or processing thread. This is the same class of issue as CVE-2020-28491 in the CBOR parser of this library, which was fixed with a LONGEST_NON_CHUNKED_BINARY guard; the asynchronous Smile parser never received the equivalent guard. No privileges or user interaction beyond the ability to submit data for parsing are required.

    Memory safety Library

    7.5 HIGH 0.4% exploit odds Fix listed
  14. CVE-2026-90484

    The MySQL client response parser in Apache bRPC dereferences a null pointer when the server peer sends a zero-payload-length response packet. A response packet whose declared payload length is zero is accepted by the frame-length check (is_full_package), after which the parser reads the response type byte via an unbounded 5-byte fetch that returns nullptr, and type = p[4] dereferences it, crashing the client process. Root Cause: In brpc::policy::MysqlReply::ConsumePartialIOBuf, buf.fetch(header, 5) returns nullptr when the buffered data only holds the 4-byte header of a zero-length packet, then the code reads p[4] (and, in the fast-auth path, p2[4]) without checking for a null or zero-length packet. Affected Scenarios: A brpc MySQL client connecting to a malicious or compromised MySQL server (or a MITM) can be crashed by a zero-length server response packet, resulting in denial of service of the client. Fix: The fix rejects zero-length response packets (returning PARSE_ERROR_ABSOLUTELY_WRONG) both for the initial response type determination and after a fast-auth marker, before reading the type byte. Upgrade to the Apache bRPC version 1.18.0 or apply the PR patch at https://github.com/apache/brpc/pull/3536.

    Memory safety Library

    6.5 MEDIUM 0.4% exploit odds Fix listed
  15. CVE-2026-125040

    A flaw was found in the SRT and WebVTT subtitle parser in GStreamer's gst-plugins-base. While sanitizing inline subtitle markup, the parser repeatedly searches the remaining text with unanchored regular expressions. A crafted subtitle cue with a long line of unmatched markup-like text can force quadratic CPU use and make a GStreamer-based application unresponsive.

    Denial of service Library

    6.5 MEDIUM 0.4% exploit odds No fix listed
  16. CVE-2025-30047

    CGM CLININET is vulnerable to Remote Code Execution caused by deserialization of untrusted data. The server exposes an HTTP-based API, used by the administrative module, that exchanges data using Java object serialization. An attacker within the same local network is able to send requests to this API and transmit a crafted serialized object. When it is deserialized, it allows arbitrary code execution with root privileges on the server.  This issue was fixed in version 2026.MS3

    Code execution Library

    9.4 CRITICAL 0.4% exploit odds Fix listed
  17. CVE-2026-91406

    Vulnerability in the Oracle Outside In Technology product of Oracle Fusion Middleware (component: Outside In PDF Export SDK). The supported version that is affected is 8.5.8. Difficult to exploit vulnerability allows unauthenticated attacker with network access via HTTP to compromise Oracle Outside In Technology. Successful attacks require human interaction from a person other than the attacker. Successful attacks of this vulnerability can result in takeover of Oracle Outside In Technology. CVSS 3.1 Base Score 7.5 (Confidentiality, Integrity and Availability impacts). CVSS Vector: (CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H).

    Other Library Oracle Outside In Technology

    7.5 HIGH 0.4% exploit odds Fix listed
  18. CVE-2026-115522

    Exposure of Resource to Wrong Sphere (CWE-668) in FileArtifactService in Google Agent Development Kit (ADK) for Python (google-adk) versions 1.19.0 before 1.39.1 and 2.0.0a1 before 2.7.0 allows a caller in one application to list, read, overwrite, or delete artifacts belonging to another application when both applications share a FileArtifactService root directory and use the same user_id (and session_id for session-scoped artifacts). FileArtifactService accepted app_name across its public API methods but omitted app_name when constructing on-disk paths (root/users/{user_id}/... instead of root/apps/{app_name}/users/{user_id}/...). Version 2.6.0 scoped writes by app_name but retained a legacy root/users/{user_id} fallback for reads and deletes; version 2.7.0 and the 1.x backport in version 1.39.1 removed the fallback and enforce full per-app path scoping.

    Disclosure Library

    5.3 MEDIUM 0.4% exploit odds Fix listed
  19. CVE-2026-116694

    OpenTelemetry Python provides the Python API and SDK for OpenTelemetry. Prior to 1.45.0, JaegerPropagator.extract() in opentelemetry-propagator-jaeger processes inbound uberctx-* baggage values without entry, per-entry byte, or total-byte limits. Services that enable the optional Jaeger propagator can perform unnecessary CPU work and short-lived heap allocations when an untrusted carrier supplies numerous or oversized values, contributing to denial of service when transport-level limits do not constrain the input. Common default Python HTTP server limits reduce the practical impact, while custom getters, non-HTTP transports, or raised header limits increase exposure. This issue is fixed in version 1.45.0.

    Denial of service Library

    6.3 MEDIUM 0.4% exploit odds Fix listed
  20. CVE-2026-128066

    Pillow is a Python imaging library. From 1.0 through 12.3.0, ImageFileDirectory_v2.load() in src/PIL/TiffImagePlugin.py retains a separate bytes object for each out-of-line IFD tag value without limiting the total data read or preventing distinct tags from referencing the same file region. A small crafted TIFF or EXIF-bearing JPEG, PNG, WebP, or MPO can therefore cause memory use quadratic in the input size. TIFF and some JPEG paths trigger parsing during Image.open(), while other formats are exposed when EXIF is requested, including through getexif() or ImageOps.exif_transpose(). The declared pixel dimensions do not bound this metadata allocation, so decompression-bomb checks do not prevent process memory exhaustion. This issue is fixed in version 13.0.0.

    Denial of service Library

    8.7 HIGH 0.4% exploit odds Fix listed
  21. CVE-2026-65982

    nginx-http-flv-module is a media streaming server based on nginx-rtmp-module. Prior to version 1.2.14, if a server received an HTTP/1.0 request without a header Host, which was permitted by HTTP core of Nginx, then r->headers_in.host was NULL, nginx-http-flv-module used r->headers_in.host->value to make a pseudo tcUrl, so NULL-pointer dereference happened. This issue has been patched in version 1.2.14.

    Memory safety Library

    8.7 HIGH 0.4% exploit odds Fix listed
  22. CVE-2026-125122

    compose-lint is a Docker Compose security linter that identifies dangerous misconfigurations before deployment. In version 0.31.0, compose-lint does not sanitize the ##[ workflow-command opener inside strings written by --format json or --format sarif to stdout. A pull request can place that opener in a service name or other Compose value, and the GitHub Actions runner recognizes it even in the middle of a serialized line when the package, pre-commit hook, or Docker image prints JSON or SARIF to the job log. Writing the output to a file and using the compose-lint GitHub Action are not affected, and the flaw does not change the lint pass or fail result. This issue is fixed in version 0.32.0.

    Other Library

    2.3 LOW 0.4% exploit odds Fix listed
  23. CVE-2026-120596

    A remotely exploitable denial-of-service vulnerability exists in the TCP options parser of the Nut/Net TCP/IP stack included in Nut/OS, developed by Ethernut/Egnite. Nut/OS releases up to version 5.1 are affected. An unauthenticated attacker can send a single TCP segment containing an unknown TCP option with a length byte of zero to any open TCP port on a vulnerable device. When processed, this malformed option triggers an infinite loop in the function NutTcpInputOptions, permanently blocking the TCP processing thread and rendering all TCP-based services unresponsive until the device is rebooted.

    Denial of service Library

    7.5 HIGH 0.4% exploit odds No fix listed
  24. CVE-2026-115572

    pypdf is a free and open-source pure-python PDF library. Prior to 6.21.2, an attacker can craft a PDF that repeatedly references the same nonexistent indirect object, causing relevant file parts such as pages to trigger repeated full-file recovery loads and long runtimes when accessed. This issue is fixed in version 6.21.2.

    Denial of service Library

    7.1 HIGH 0.4% exploit odds Fix listed
  25. CVE-2026-116571

    FreeSWITCH is a Software Defined Telecom Stack enabling the digital transformation from proprietary telecom switches to a software implementation that runs on any commodity hardware. Prior to 1.11.2, rtmp_rtp2rtmpH264() built an AVC decoder configuration record in a fixed stack buffer by copying wire-derived SPS and PPS values without checking that their sizes and framing fit. An established H.264 call bridged toward an RTMP endpoint could supply oversized SPS and PPS parameter sets and overrun the function's stack frame. The flaw affects only deployments that build and load mod_rtmp and bridge RTP media into an RTMP leg; the module is disabled in the stock build and configuration. The corruption can abort FreeSWITCH and terminate every active call, and builds without stack protection receive a limited stack-write primitive, although code execution has not been demonstrated. This issue is fixed in version 1.11.2.

    Memory safety Library Freeswitch

    5.9 MEDIUM 0.4% exploit odds Fix listed
  26. CVE-2026-120752

    PyMuPDF 1.25.3 before 2.0 contains a stack-based buffer overflow vulnerability in JM_set_annot_callout_line() that copies a caller-supplied number of callout points into a fixed three-element stack array. Attackers who control the callout argument passed to Page.add_freetext_annot() can supply four or more points to overwrite the stack frame, crashing the process or executing code.

    Memory safety Library

    7.7 HIGH 0.4% exploit odds Fix listed
  27. CVE-2026-96643

    Pillow is a Python imaging library. From 11.0.0 until 13.0.0, Exif.load() in src/PIL/Image.py repeatedly slices the remaining EXIF data while it begins with the six-byte Exif header. src/PIL/PngImagePlugin.py accepts an uncapped PNG eXIf chunk, so a file filled with repeated headers causes a new shrinking copy on every iteration and quadratic CPU work. The condition is reachable through Image.getexif(), Image._getexif(), and ImageOps.exif_transpose(), including common photo-upload rotation workflows. This issue is fixed in version 13.0.0.

    Other Library Python Pillow

    7.1 HIGH 0.4% exploit odds Fix listed
  28. CVE-2026-96637

    Pillow is a Python imaging library. Prior to 13.0.0, SpiderImageFile.seek() in src/PIL/SpiderImagePlugin.py reparses a later SPIDER stack frame and replaces self._size without rerunning Image._decompression_bomb_check(). A tiny file can use a safe first frame to pass Image.open() and then declare very large dimensions in a later mode F frame. Loading that frame allocates approximately four bytes per attacker-declared pixel and can consume multiple gigabytes of memory. This issue is fixed in version 13.0.0.

    Denial of service Library Python Pillow

    7.1 HIGH 0.4% exploit odds Fix listed
  29. CVE-2026-123805

    SIPSorcery is a WebRTC, SIP, and VoIP library for C# and .NET. Prior to 10.0.14, SctpSackChunk.ParseChunk reads the numGapAckBlocks and numDuplicateTSNs fields from an attacker-controlled SCTP SACK chunk without checking that the corresponding entries fit the declared chunk length. A negotiated WebRTC peer can send a checksum-valid malformed SACK chunk that drives the parser beyond the 262144-byte receive buffer. The resulting IndexOutOfRangeException bypasses the recoverable ApplicationException handler in RTCSctpTransport.DoReceive, causing the generic handler to exit the dedicated receive loop and permanently terminate the SCTP association and all data channels. This issue is fixed in 10.0.14.

    Memory safety Library

    7.5 HIGH 0.4% exploit odds Fix listed
  30. CVE-2026-128237

    Scapy before 2.8.0 contains a denial of service vulnerability in BER_num_dec() and BERcodec_OID.do_dec() that allows remote attackers to exhaust CPU by supplying oversized OID arcs. Attackers can send ASN.1/BER data with many base-128 continuation bytes to cause quadratic CPU consumption or an uncaught ValueError crash on Python 3.11+.

    Denial of service Library

    6.9 MEDIUM 0.4% exploit odds Fix listed
  31. CVE-2026-127208

    datamodel-code-generator before 0.84.0 contains a code injection vulnerability that allows attackers to achieve arbitrary code execution by supplying a malicious JSON Schema with a crafted x-python-type annotation containing a string ast.Constant node in a subscript slice position, which bypasses validation in parse_python_type_annotation and _python_type_expr_from_ast. Attackers can embed malicious Python expressions such as __import__ calls within quoted string literals in the generated source code, which Pydantic v2 evaluates as a ForwardRef via eval() during model_rebuild(), resulting in arbitrary code execution on developer workstations, CI/CD pipelines, or applications that generate and import the produced models at runtime.

    Injection Library

    8.6 HIGH 0.4% exploit odds Fix listed
  32. CVE-2026-115456

    A flaw was found in gcc. An attacker could cause a Denial of Service (DoS) by passing a string that is not null-terminated to the std::from_chars() function in the libstdc++ library. This issue triggers an out-of-bounds read beyond the allocated buffer, potentially leading to invalid data parsing or an application crash.

    Memory safety Library

    3.7 LOW 0.4% exploit odds No fix listed
  33. CVE-2026-101507

    Vulnerability in Oracle Trusted Answer Search (component: PLSQL package). The supported version that is affected is 26.1. Easily exploitable vulnerability allows low privileged attacker with network access via HTTP to compromise Oracle Trusted Answer Search. Successful attacks of this vulnerability can result in unauthorized access to critical data or complete access to all Oracle Trusted Answer Search accessible data. CVSS 3.1 Base Score 6.5 (Confidentiality impacts). CVSS Vector: (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N).

    Other Library

    6.5 MEDIUM 0.4% exploit odds No fix listed
  34. CVE-2026-122310

    Kyverno from 1.19.1 before 1.19.2 validates a Policy's context.apiCall urlPath after cleaning it but executes the raw path, whose fragment or query the Kubernetes client discards. A namespace-scoped user who can create a Policy can hide ../ segments behind # or ? to reach cluster-scoped endpoints with Kyverno's ServiceAccount, gaining cluster admin.

    Auth & access Library Kyverno

    9.4 CRITICAL 0.4% exploit odds Fix listed
  35. CVE-2026-114501

    Pillow is a Python imaging library. From 11.2.1 until 13.0.0, Arrow export assumes that an image's rows are tightly contiguous and exports from im->blocks[0].ptr even when PILLOW_ALIGNMENT makes aligned_linesize include row padding or an aligned offset before the first row. The resulting Arrow buffer can include uninitialized padding, bytes preceding the first row, or stale heap contents from previously freed images. Applications that configure alignment and export images to Arrow can therefore disclose one user's pixels or other process memory to another user. This issue is fixed in version 13.0.0.

    Other Library Python Pillow

    6.3 MEDIUM 0.4% exploit odds Fix listed
  36. CVE-2026-122126

    reSIProcate reTurn client through 1.14.0 contains a heap buffer overflow in TurnTcpSocket::handleReadHeader(), which TurnTlsSocket inherits, that allows malicious TURN servers or on-path attackers to overwrite memory via oversized frame lengths. Attackers can send STUN or ChannelData frames declaring up to 65535 bytes, writing roughly 57 KB past the fixed 8192-byte mReadBuffer and crashing the client. Only the synchronous TCP and TLS client is affected; the asynchronous client (TurnAsyncSocket) and the reTurn server are not.

    Memory safety Library

    8.2 HIGH 0.4% exploit odds Fix listed
  37. CVE-2026-114459

    A flaw was found in mod_qos. When inspecting HTTP response bodies, improper boundary checking can cause the module to read memory outside allocated buffer boundaries. An unauthenticated remote attacker can exploit this vulnerability by triggering specifically crafted server responses, causing an Apache worker process to crash and resulting in a Denial of Service (DoS).

    Memory safety Library

    3.7 LOW 0.4% exploit odds No fix listed
  38. CVE-2026-11316

    A out-of-bounds write vulnerability exists in the dr_wav drwav__read_cue_to_metadata_obj functionality of dr_libs (version(s): 0.14.5). A specially crafted .wav file can lead to an out-of-bounds write. An attacker can provide a malicious file to trigger this vulnerability.

    Memory safety Library

    8.8 HIGH 0.4% exploit odds No fix listed
  39. CVE-2026-123584

    The hickory-net crate in Hickory DNS 0.26.0 before 0.26.2 stops listening and returns an error when a client UDP connection receives a datagram that cannot be parsed as a DNS message. Remote attackers who guess the query's source port, without knowing the transaction ID, can spoof malformed responses to abort client queries, causing denial of service.

    Denial of service Library

    8.2 HIGH 0.4% exploit odds Fix listed
  40. CVE-2026-111770

    The Model Context Protocol TypeScript SDK provides TypeScript client and server implementations of the Model Context Protocol. From 2.0.0 until 2.3.0 in @modelcontextprotocol/server, applications that shared one stateless HTTP transport or one McpServer or Server instance among multiple clients could deliver responses and notifications to the wrong client. The affected surface is an MCP server reached over HTTP by multiple clients. The trigger is reuse of one stateless transport or server instance. Messages were associated with a connection using only the JSON-RPC request identifier. SDK clients start request identifiers at zero, so collisions occur in normal operation. Cross-client delivery exposes one client's responses and notifications to another client. The issue requires an HTTP endpoint reachable by more than one client and an application that constructs the server or transport once at startup instead of once per request or session. stdio servers and factory-created per-call servers are not affected. Upgrading causes reused instances to fail and therefore requires an application code change. This issue is fixed in @modelcontextprotocol/server 2.3.0.

    Other Library

    7.7 HIGH 0.4% exploit odds Fix listed
  41. CVE-2026-74759

    Google::Auth versions 0.12 for Perl allow server side request forgery via unvalidated credential_source URLs in AWS external account credentials. An external_account configuration with the AWS subject token type selects Google::Auth::ExternalAccountCredentials::Aws. Its retrieve_subject_token sends requests to the imdsv2_session_token_url, url and region_url fields of credential_source without the host check that the other credential classes apply. Any caller that builds credentials from a configuration it does not fully control issues GET and PUT requests to hosts the configuration names, including internal services and link-local metadata endpoints.

    Disclosure Library

    9.8 CRITICAL 0.4% exploit odds No fix listed
  42. CVE-2026-106070

    A flaw in the Node.js `zlib.ZipFile` API can cause it to return bytes that do not belong to the requested entry after the local and central header validation fails. When `zipEntry.modified` is read on an archive whose local header does not match its central directory, either directly or while compacting the archive, the validation error is caught but the parsed local header is left cached without a content offset. A later content read then returns attacker-controlled bytes from the start of the file instead of rejecting the archive, and CRC-32 verification still passes because the attacker controls the central directory CRC. A Node.js-based scanner can therefore approve different content from what another ZIP implementation extracts from the same archive. This vulnerability affects Node.js **26.x**.

    Other Library

    3.7 LOW 0.4% exploit odds No fix listed
  43. CVE-2026-125111

    Rack is a modular Ruby web server interface. Prior to 2.2.25, 3.1.23, and 3.2.8, Rack::Multipart::Parser can retain parsed data in its internal StringScanner buffer when reads end inside or immediately after multipart headers. An unauthenticated attacker can align many small parts with rack.multipart.buffer_size boundaries so the buffer grows with the request despite per-field, header, and preamble limits, causing a worker to exhaust memory. The attack requires only a well-formed multipart body and affects applications that parse multipart requests. This issue is fixed in versions 2.2.25, 3.1.23, and 3.2.8.

    Denial of service Library

    7.5 HIGH 0.4% exploit odds Fix listed
  44. CVE-2026-112064

    The backend module for reviewing the consent history accepted sorting, grouping and index options from the request and used them in a database query without validation. Only the filter conditions of that query were bound as parameters. A backend user with access to the history or statistics functions could use this to read arbitrary data from the database, including password hashes of backend and frontend users.

    Injection Library

    7.1 HIGH 0.4% exploit odds No fix listed
  45. CVE-2026-123864

    A re-entrancy flaw in Node.js HTTPS lets an unauthenticated client read memory belonging to other connections, including TLS private keys and other users' requests. TLSWrap::DoWrite() calls ClearOut() directly when asked to write zero bytes; if that write is issued from JavaScript running inside the HTTP parser's on_body callback, ClearOut() decrypts the next TLS record while the parser is still mid-body, re-entering llhttp_execute() and causing a use-after-free and an out-of-bounds heap read. The server then copies that freed or unrelated heap memory into the response for the request body. Triggering it requires a handler that calls res.flushHeaders() before the request body ends and later calls res.end() with a non-empty chunk, and reading the leaked memory requires any ordinary route that echoes or stores a request body. Cleartext node:http is not affected. This vulnerability affects Node.js 26.x, 24.x, and 22.x.

    Memory safety Library

    8.1 HIGH 0.4% exploit odds No fix listed
  46. CVE-2026-118769

    Botan is a C++ cryptography library. From 2.1.0 until 3.13.0, the Botan C API function botan_block_cipher_block_size converted a size_t block size to int without a bounds check. In an application that accepts external algorithm parameter strings and uses the FFI raw ECB mode, an attacker can select the variable-block-size Lion cipher so that the reported size is truncated, the caller allocates insufficient memory, and later block processing causes memory corruption. The C++ API and normally used C API modes such as CBC and CTR are unaffected. This issue is fixed in version 3.13.0.

    Memory safety Library

    9.2 CRITICAL 0.4% exploit odds Fix listed
  47. CVE-2026-116684

    The Model Context Protocol Python SDK is the official Python SDK for Model Context Protocol servers and clients. From 1.10.0 until 1.30.0 and 2.2.0, server-side bearer authentication configured through AuthSettings with token_verifier accepts a valid token issued for another service when the authorization server serves multiple resources and the TokenVerifier does not independently validate the aud audience. This resource-confusion flaw lets a token holder authenticate to the wrong MCP server and exercise that token's resulting access. This issue is fixed in versions 1.30.0 and 2.2.0.

    Auth & access Library

    6.8 MEDIUM 0.4% exploit odds Fix listed
  48. CVE-2026-96638

    Pillow is a Python imaging library. From 9.1.0 until 13.0.0, BmpRleDecoder.decode() in src/PIL/BmpImagePlugin.py permits crafted RLE BMP delta escapes and absolute-mode runs to append beyond the remaining dest_length without a per-branch cap. A delta can add roughly 255 image rows in one iteration, so a kilobyte-sized file with a large but permitted width can expand the data bytearray to hundreds of megabytes or more, causing memory-exhaustion denial of service. This issue is fixed in version 13.0.0.

    Denial of service Library Python Pillow

    6.5 MEDIUM 0.4% exploit odds Fix listed
  49. CVE-2026-96641

    Pillow is a Python imaging library. Prior to 13.0.0, PILfont text-width accumulation in src/_imaging.c checks positive glyph advances against INT_MAX but does not check negative advances against INT_MIN. Repeating a signed 16-bit advance of -32768 can therefore execute a signed C integer overflow in xsize += dx, producing a large positive width from a negative mathematical result. Applications that use getlength(), getsize(), or derived layout, quota, serialization, or allocation calculations can experience logic-integrity errors or resource consumption, while the high-level mask path commonly retains a decompression-bomb check. This issue is fixed in version 13.0.0.

    Memory safety Library Python Pillow

    6.5 MEDIUM 0.4% exploit odds Fix listed
  50. CVE-2026-96636

    Pillow is a Python imaging library. Prior to 13.0.0, MpoImageFile.seek() and DcxImageFile.seek() reparse later MPO or DCX frames and replace self._size without rerunning Image._decompression_bomb_check(). A crafted multi-frame image can use a tiny first frame to pass Image.open() and then declare an over-limit later frame. Normal frame iteration through seek(), load(), or ImageSequence.Iterator can decode that frame and commit hundreds of megabytes or more, causing memory-exhaustion denial of service. This issue is fixed in version 13.0.0.

    Denial of service Library Python Pillow

    6.5 MEDIUM 0.4% exploit odds Fix listed

CVEs that affect library, sorted by how likely they are to be exploited (EPSS).