Red Hat Linux Security Advisories & CVEs
4742 advisories tracked · Red Hat Security Data API · direct feeds checked every minute; rate-limited backstops use a safe source cadence
Android app · Google Play
Monitor Red Hat CVEs from your phone.
Choose a whole vendor or a precise platform, then receive matching security advisories by phone notification, email, or both. Coverage follows 34 official vendor sources and 160+ reviewed platform categories.
Security advisories for your Red Hat release
Pick your distribution release to see every advisory issued for it and its severity mix. Fixes ship as errata — keep the system patched. This is the release's advisory history, not a per-package scan.
Official source
Red Hat Security Data API
Red Hat Enterprise Linux errata (RHSA) via the official Red Hat Security Data API — CVE severity, CVSS and affected packages. A credential-free official source.
Latest Red Hat advisories
High [CVE-2026-54900] Heap corruption via crafted JSON object key
Oj (Optimized JSON) is a JSON parser and Object marshaller packaged as a Ruby gem. In versions prior to 3.17.2, when in usual mode with create_id enabled, Oj::Parser#parse is vulnerable to heap corruption via a negative-size memcpy. When a JSON object key is exactly 65,535 bytes long, an integer truncation in form_attr (usual.c:63) converts the length to -1 before passing it to memcpy. This causes memcpy to copy SIZE_MAX bytes (interpreted as a huge size_t), corrupting heap memory and crashing the process. The issue has been fixed in version 3.17.2. A flaw was found in Oj, a Ruby gem for JSON parsing. When processing a specially crafted JSON object key of a specific length (65,535 bytes) in usual mode with create_id enabled, an integer truncation vulnerability occurs. This leads to a negative size being passed to the `memcpy` function, causing heap memory corruption and crashing the process. A remote attacker could exploit this to cause a denial of service. Red Hat's only product shipping the Oj Ruby gem (Compliance Backend) already includes version 3.17.3, which contains the fix for this vulnerability. No Red Hat products are affected. Red Hat severity: Important — CVSS 7.5 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H). Weakness: CWE-131.
High [CVE-2026-54897] Use-After-Free in Oj::Doc Iterators via reentrant close
Oj (Optimized JSON) is a JSON parser and Object marshaller packaged as a Ruby gem. Prior to 3.17.2, Oj::Doc iterators (each_value, each_child, each_leaf) were vulnerable to a heap use-after-free. When a Ruby block yielded during iteration calls doc.close or d.close, the document's heap memory is freed while the C iterator is still running. When control returns from the block, the iterator reads from the freed region, producing a use-after-free accessible from pure Ruby. This issue has been fixed in version 3.17.2. A flaw was found in Oj (Optimized JSON), a Ruby gem for parsing JSON. An attacker could exploit this when a Ruby code block, executed during iteration, prematurely closes the document. This action frees the document's memory while the iterator is still active, leading to an attempt to access freed memory and potentially causing application instability or other unpredictable behavior. Red Hat's only product shipping the Oj Ruby gem (Compliance Backend) already includes version 3.17.3, which contains the fix for this vulnerability. No Red Hat products are affected. Red Hat severity: Important — CVSS 7.8 (CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H). Weakness: CWE-364.
High [CVE-2026-54896] Heap buffer overflow in exception serialization
Oj (Optimized JSON) is a JSON parser and Object marshaller packaged as a Ruby gem. In versions prior to 3.17.2, when in object mode, Oj.dump is vulnerable to a heap buffer overflow when serializing Exception objects with a large:indent value. The serializer allocates a buffer sized for the object's attributes but does not account for the indent bytes added on each write. With indent: 5000, the accumulation of 5,000-byte indent strings overflows the 13,150-byte heap allocation, corrupting adjacent heap memory. This issue has been fixed in version 3.17.2. A flaw was found in Oj (Optimized JSON), a Ruby gem used for processing JSON data. This vulnerability occurs when the software attempts to serialize (convert into a format for storage or transmission) Exception objects with a very large indentation setting. The internal process for handling this serialization does not properly manage memory, leading to a heap buffer overflow. This memory corruption could allow an attacker to disrupt the application, causing a denial of service, or potentially execute unauthorized code. Red Hat's only product shipping the Oj Ruby gem (Compliance Backend) already includes version 3.17.3, which contains the fix for this vulnerability. No Red Hat products are affected. Red Hat severity: Important — CVSS 7.8 (CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H). Weakness: CWE-787.
High [CVE-2026-54592] Denial of Service via deeply nested JSON input
Oj (Optimized JSON) is a JSON parser and Object marshaller packaged as a Ruby gem. In versions prior to 3.17.3, Oj::Doc#each_child, when invoked recursively over a deeply nested JSON document, overflows a fixed-size stack buffer and aborts the process, leading to DoS. In a two-step chain in ext/oj/fast.c, doc_each_child increments doc->where past the where_path[MAX_STACK = 100] array with no bounds check and never restores it (the doc->where-- is missing), so calling each_child recursively from inside the yield block drives doc->where beyond the array. On the next entry the function copies the path into the 800-byte stack-local buffer save_path[MAX_STACK] using wlen = doc->where - doc->where_path, so when the previous recursive call left doc->where past where_path[100] the wlen exceeds MAX_STACK and the memcpy overflows save_path on the C stack; because the Oj::Doc parser imposes no JSON nesting-depth limit (relying on a C-stack pressure check), deeply nested attacker input reaches this path. This issue has been fixed in version 3.17.3. A flaw was found in Oj, a Ruby gem for JSON parsing. A remote attacker can exploit this vulnerability by providing a deeply nested JSON document. This can lead to a stack buffer overflow in the `Oj::Doc#each_child` function, causing the process to abort and resulting in a denial of service (DoS) for the affected system.
High [CVE-2026-55223] Remote code execution via deserialization vulnerability
c3p0 is a JDBC Connection pooling library. In versions prior to 0.14.0, c3p0 in combination with other libraries, can compose to a "sink" for deserialization gadgets. The JDBC spec's DataSource.getConnection() and ConnectionPoolDataSource.getPooledConnection() match the getXXX() form, so JavaBean libraries treat them as "properties" assumed safe while they actually call into JDBC drivers. Attackers can thus craft malicious DataSource objects whose property lookups invoke vulnerable drivers, then smuggle them in serialized form to where an application deserializes and auto-resolves bean properties — triggering the attack. This requires a susceptible DataSource/ConnectionPoolDataSource and JDBC driver on the CLASSPATH, plus a carrier that auto-looks-up JavaBean properties on = deserialization, most commonly a collection paired with an Apache commons-beanutils Comparator that sorts by bean properties. c3p0 supplied that susceptible DataSource/ConnectionPoolDataSource, which was an essential component of the trigger. This issue has been fixed in version 0.14.0. This vulnerability allows a remote attacker to potentially execute arbitrary code by crafting a malicious data source object. This requires specific conditions, including the presence of a susceptible JDBC driver and a mechanism for automatic property resolution during deserialization.
High [CVE-2026-54672] Arbitrary code execution through AppImage library loading vulnerability
electron-updater allows for automatic updates for Electron apps. Prior to 26.15.0, AppImage targets built by app-builder-lib could use an empty path component when setting the LD_LIBRARY_PATH environment variable at runtime. This causes the current working directory to be added to the dynamic linker search path, which may allow an attacker to execute arbitrary code by placing a malicious shared library in the directory from which the AppImage is launched. This issue has been fixed in version 26.15.0. This could allow the attacker to run arbitrary code on the affected system. The vulnerability in electron-builder's AppImage packaging only affects applications distributed as AppImage format, where an empty LD_LIBRARY_PATH component causes the current working directory to be added to the dynamic linker search path. Red Hat products that bundle electron-builder components are not affected: the goose RPM is distributed as an RPM package (not AppImage), and Podman Desktop uses Flatpak and tar.gz targets on Linux (AppImage is not configured in its electron-builder config). The vulnerable AppImage code path is never exercised in any Red Hat product. Red Hat severity: Important — CVSS 7.3 (CVSS:3.1/AV:L/AC:L/PR:L/UI:R/S:U/C:H/I:H/A:H). Weakness: CWE-427. Red Hat lists Red Hat Build of Podman Desktop; Red Hat Enterprise Linux 10; Red Hat Enterprise Linux 9 as not affected.
High [CVE-2026-58014] off-by-one error in glib/gkeyfile.c via "g_key_file_get_locale_string_list"
A flaw was found in GLib. An off-by-one error can occur in the g_key_file_get_locale_string_list function in the gkeyfile.c file when loading a key file with an empty value. This flaw can cause an out-of-bounds access of 1 byte or a denial of service when the out-of-bounds access crosses a page boundary. Any applications loading or parsing untrusted key files, typically.desktop or.ini files, are vulnerable to this issue. Due to these reasons, this vulnerability has been rated with a moderate severity. Red Hat severity: Moderate — CVSS 7.3 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L). Weakness: CWE-193. Affected products named by the advisory: Red Hat Enterprise Linux 10.0 Extended Update Support; Red Hat Enterprise Linux 7 Extended Lifecycle Support; Red Hat Enterprise Linux 8.4 Advanced Mission Critical Update Support; Red Hat Enterprise Linux 8.4 Extended Update Support Long-Life Add-On; and 22 more. Affected products named by the advisory: Red Hat Enterprise Linux 8.6 Advanced Mission Critical Update Support; Red Hat Enterprise Linux 8.6 Extended Update Support Long-Life Add-On; Red Hat Enterprise Linux 8.8 Telecommunications Update Service; Red Hat Enterprise Linux 8.8 Update Services for SAP Solutions; and 18 more.
High [CVE-2026-58374] Denial of Service via malformed Wi-Fi 7 Multi-Link Operation association request
In hostapd before 2.12, a missing bounds check in AP-mode Wi-Fi 7 (IEEE 802.11be) Multi-Link Operation (MLO) association request processing allows an unauthenticated attacker within wireless range to send a crafted management frame containing a malformed Multi-Link Element or Per-STA Profile subelement. In hostapd_process_ml_assoc_req() in src/ap/ieee802_11_eht.c, the received link_id field can be parsed as value 15, but the corresponding links[] storage only has valid entries for lower link IDs (0 through 14). This causes an out-of-bounds write / small memory corruption during association processing before the 4-way handshake. The attack does not require network credentials, prior authentication, or user interaction. The confirmed practical impact is denial of service through hostapd process termination. This affects hostapd v2.11 and newer development snapshots before v2.12 when built with CONFIG_IEEE80211BE enabled. The issue is fixed in hostapd v2.12 and the upstream 2026-1 fixes. A flaw was found in hostapd. Red Hat severity: Moderate — CVSS 7.1 (CVSS:3.1/AV:A/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:H). Weakness: CWE-787. Affected Red Hat products: Red Hat Enterprise Linux 10; Red Hat Enterprise Linux 8; Red Hat Enterprise Linux 9. Red Hat does not currently list a fixing RHSA for this CVE. Affected products named by the advisory: Red Hat package: hostapd.
High [CVE-2026-53433] Denial of Service via inefficient HTTP body processing
fzf is vulnerable to a Denial of Service (DoS) due to inefficient HTTP body processing in the --listen mode due to inefficient HTTP body processing using repeated string concatenation, resulting in quadratic time complexity (O(n²)). A crafted POST request with many small segments can trigger excessive CPU usage during request handling. This allows a single malicious request to monopolize the single‑threaded HTTP server, blocking all other clients and resulting in denial of service. This issue was fixed in version 0.73.1. A flaw was found in fzf, a command-line fuzzy finder. This vulnerability allows a remote attacker to cause a Denial of Service (DoS) by sending a crafted POST request with many small segments to the `--listen` mode. The inefficient HTTP body processing, which uses repeated string concatenation, leads to quadratic time complexity, causing excessive CPU usage and blocking other clients. Red Hat has corrected the impact from IMPORTANT to MODERATE — the original AI-Bot CVSS of 7.5 (AV:N/AC:L) did not account for the --listen mode being a non-default feature that must be explicitly enabled. The CVEORG CVSSv4 score of 5.7 (AV:L/AT:P) more accurately reflects the prerequisite nature of the attack. Weakness: CWE-1046.
High [CVE-2026-49434] Unauthorized broker instantiation via improper input validation in LDAP entries
Improper Input Validation vulnerability in Apache ActiveMQ Broker, Apache ActiveMQ, Apache ActiveMQ All. An attacker that has access to publish or modify entries in LDAP that match the configured searchBase and searchFilter can instantiate denied transports inside the broker JVM. This can be used to fetch an attacker URL and spawn a second BrokerService inside the same JVM. Users are recommended to upgrade to version 6.2.7 or 5.19.8, which fixes the issue. An attacker with privileges to publish or modify entries in Lightweight Directory Access Protocol (LDAP) can exploit an improper input validation vulnerability. Red Hat products ship Apache ActiveMQ Classic components as transitive dependencies. The vulnerability is in the LdapNetworkConnector feature specific to Classic ActiveMQ, which allows an attacker with LDAP write access to instantiate denied transports and spawn a rogue broker. This feature is not configured or used in any Red Hat product deployment. Red Hat AMQ Broker is based on Apache ActiveMQ Artemis, a separate codebase that does not include the LdapNetworkConnector. Red Hat severity: Important — CVSS 7.6 (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:H). Weakness: CWE-90. Affected products named by the advisory: Red Hat AMQ Broker 7; Red Hat AMQ Clients; Red Hat build of Apache Camel for Spring Boot 4; Red Hat Data Grid 8; and 7 more.
High [CVE-2026-49432] Denial of Service via improper input validation in STOMP connector
Improper Input Validation vulnerability in Apache ActiveMQ, Apache ActiveMQ All, Apache ActiveMQ Stomp. A remote unauthenticated peer that can reach an exposed STOMP connector can trigger denial-of-service behavior by sending a negative content-length. For the NIO STOMP transport, an attacker can keep streaming body bytes and grow the per-connection command buffer beyond configured limits to cause OOM. For the blocking STOMP protocol, an error will instead force abnormal transport exception handling for the affected connection and closure. Users are recommended to upgrade to version 6.2.7 or 5.19.8, which fixes the issue. This can lead to a denial of service (DoS) condition, either by consuming excessive memory and causing an Out-Of-Memory (OOM) error or by forcing the abnormal closure of affected connections. Red Hat products ship Apache ActiveMQ Classic components as transitive dependencies. The vulnerability is in the Classic ActiveMQ STOMP connector's content-length validation, allowing an unauthenticated attacker to cause OOM via a negative content-length value. Apache ActiveMQ Artemis, which powers Red Hat AMQ Broker, has its own STOMP protocol implementation (artemis-stomp-protocol) that does not share this code path. The Classic STOMP connector is not exposed in Red Hat product deployments.
High [CVE-2026-50734] Denial of Service via crafted WireFormatInfo frame
Memory Allocation with Excessive Size Value vulnerability in Apache ActiveMQ Client, Apache ActiveMQ, Apache ActiveMQ All. An unauthenticated network attacker can cause a broker DoS by sending a crafted WireFormatInfo frame with a malicious large size value. The value is not validate and causes the broker to attempt allocation during pre-auth negotiation which can trigger OOM and crash the broker. Users are recommended to upgrade to version 6.2.7 or 5.19.8, which fixes the issue. Consequently, this can lead to an Out Of Memory (OOM) error, resulting in a Denial of Service (DoS) and crashing the broker. A flaw was found in Apache ActiveMQ Classic's OpenWire protocol handling. Red Hat severity: Important — CVSS 7.5 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H). Weakness: CWE-770. Affected Red Hat products: AMQ Clients 2026.Q3; Red Hat AMQ Broker 7.13.6; Red Hat AMQ Broker 7.14.1; Red Hat AMQ Broker 7; Red Hat AMQ Clients; Red Hat Fuse 7; Red Hat Single Sign-On 7. Red Hat lists Red Hat build of Apache Camel for Spring Boot 4; Red Hat Data Grid 8; Red Hat Enterprise Linux 8; Red Hat JBoss Enterprise Application Platform 7; Red Hat JBoss Enterprise Application Platform 8; Red Hat Satellite 6 as not affected. Will not fix / out of support: Red Hat Fuse 7. Red Hat fixing advisory: RHSA-2026:69459, RHSA-2026:66545, RHSA-2026:66488.
High [CVE-2026-50750] Denial of Service via repeated BrokerInfo commands
Denial of Service via Out of Memory vulnerability in Apache ActiveMQ Broker, Apache ActiveMQ, Apache ActiveMQ All. Following the fix for CVE-2026-49270 an unauthenticated attacker can now cause broker OOM by sending an repeated BrokerInfo commands without sending a ConnectionInfo, until the broker will crash with OOM. Users are recommended to upgrade to version 6.2.7, which fixes the issue. This can lead to an Out of Memory condition, causing the broker to crash and resulting in a Denial of Service. Red Hat products that include Apache ActiveMQ classic components ship versions prior to 5.19.7 (5.x line) and prior to 6.2.6 (6.x line). The vulnerable code was introduced as a regression in versions 5.19.7 and 6.2.6 while fixing CVE-2026-49270, and is not present in the versions shipped by Red Hat. Products shipping Apache ActiveMQ Artemis are not affected as Artemis is a separate codebase. Red Hat severity: Important — CVSS 7.5 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H). Weakness: CWE-770. Red Hat lists Red Hat AMQ Broker 7; Red Hat AMQ Clients; Red Hat build of Apache Camel for Spring Boot 4; Red Hat Data Grid 8; Red Hat Enterprise Linux 8; Red Hat Fuse 7; Red Hat JBoss Enterprise Application Platform 7; Red Hat JBoss Enterprise Application Platform 8; Red Hat Satellite 6; Red Hat Single Sign-On 7 as not affected.
High [CVE-2026-53916] Apache ActiveMQ, Apache ActiveMQ All, Apache ActiveMQ Stomp: Unbounded header buffer in STOMP NIO codec
Memory Allocation with Excessive Size Value vulnerability in Apache ActiveMQ, Apache ActiveMQ All, Apache ActiveMQ Stomp. An unauthenticated client that opens a STOMP NIO connection can send header bytes that never terminate which makes the broker buffer them without limit, exhausting the JVM heap. Users are recommended to upgrade to version 6.2.7 or 5.19.8, which fixes the issue. Red Hat products ship Apache ActiveMQ Classic components as transitive dependencies. The vulnerability is in the Classic ActiveMQ STOMP NIO codec, which does not limit the size of header byte buffering, allowing an unauthenticated attacker to exhaust JVM heap memory. Apache ActiveMQ Artemis, which powers Red Hat AMQ Broker, has its own STOMP protocol implementation (artemis-stomp-protocol) that does not share this code path. The Classic STOMP NIO codec is not exposed in Red Hat product deployments. Red Hat severity: Important — CVSS 7.5 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H). Weakness: CWE-789. Affected Red Hat products: Red Hat AMQ Broker 7.13.6; Red Hat AMQ Broker 7.14.1; Red Hat AMQ Broker 7; Red Hat AMQ Clients; Red Hat build of Apache Camel for Spring Boot 4; Red Hat Data Grid 8; Red Hat Enterprise Linux 8; Red Hat Fuse 7; Red Hat JBoss Enterprise Application Platform 7; Red Hat JBoss Enterprise Application Platform 8; Red Hat Satellite 6; Red Hat Single Sign-On 7.
High [CVE-2026-54475] Information disclosure due to broken temporary destination isolation
Missing Authorization vulnerability in Apache ActiveMQ Broker, Apache ActiveMQ All, Apache ActiveMQ. Apache ActiveMQ Classic temporary destinations are expected to be isolated to the connection that created them. The isolation can be broken as this is only checked in the client, allowing a different connection to consume from another connection's temporary destination. Users are recommended to upgrade to version 6.2.7, which fixes the issue. Temporary destinations, which are designed to be private to a specific connection, can be accessed by other connections due to a missing authorization check. This allows an unauthorized connection to consume messages from another connection's temporary destination, leading to information disclosure. Red Hat products ship Apache ActiveMQ Classic components as transitive dependencies. Red Hat AMQ Broker is based on Apache ActiveMQ Artemis, which has its own temporary destination implementation. The Classic broker's temporary destination logic is not exercised at runtime in Red Hat product deployments. Red Hat severity: Important — CVSS 8.2 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:L). Weakness: CWE-1220. Affected Red Hat products: Red Hat Fuse 7. Red Hat lists Red Hat AMQ Broker 7; Red Hat Data Grid 8; Red Hat Enterprise Linux 8; Red Hat JBoss Enterprise Application Platform Expansion Pack as not affected.
High [CVE-2026-13149] Denial of Service due to exponential-time complexity
brace-expansion through 5.0.6 is vulnerable to denial of service. The expand() function exhibits exponential-time complexity in the number of consecutive non-expanding '{}' brace groups. An attacker who passes a crafted string to expand(), directly or transitively, can cause significant CPU consumption and event-loop blocking. The max option does not mitigate this, as it bounds the output size rather than the recursion work. An attacker can exploit a vulnerability in the `expand()` function by providing a specially crafted string. This can result in a Denial of Service (DoS) for the affected system. A flaw was found in brace-expansion, a widely-used npm package for expanding brace sequences. Red Hat severity: Important — CVSS 7.5 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H). Weakness: CWE-1333. Affected products named by the advisory: Cryostat 4 on RHEL 9; Red Hat AMQ Broker 7.13.6; Red Hat AMQ Broker 7.14.1; Red Hat Enterprise Linux 10.0 Extended Update Support; and 48 more. Affected products named by the advisory: Red Hat Enterprise Linux 8; Red Hat Enterprise Linux 9.6 Extended Update Support; Red Hat Advanced Cluster Security 4.9; Red Hat Advanced Cluster Security for Kubernetes 4.10; and 44 more.
High [CVE-2026-45822] Denial of Service via crafted input
decode-uri-component through 0.4.1 is vulnerable to denial of service. The decode() function splits input on '%' producing N tokens and calls decodeComponents(), exhibiting super-linear parsing time: 200 '%ab' tokens takes approximately 0.7s, 700 tokens approximately 6s, and 1400 tokens approximately 33s. An attacker can cause significant CPU consumption and event-loop blocking via crafted input. A flaw was found in the `decode-uri-component` library. This vulnerability allows a remote attacker to trigger a Denial of Service (DoS) by submitting specially crafted input. The `decode()` function, when processing a large number of encoded URI components, consumes excessive CPU resources, which can lead to the application becoming unresponsive and unavailable. In Red Hat products where this package is bundled (OpenShift Console, Quay, Pipelines, RHOAI, and others), exploitation requires that attacker-controlled input containing crafted percent-encoded strings reaches the decode() function without prior length validation. Red Hat rates this as Moderate severity since the impact is limited to availability with no confidentiality or integrity impact, consistent with the CNA's CVSS 4.0 assessment of 6.6 Medium. Weakness: CWE-1050. Affected products named by the advisory: Red Hat Migration Toolkit 1.8; Red Hat Quay 3.10; Red Hat Quay 3.12; Red Hat Quay 3.15; and 7 more.
High [CVE-2026-12243] Information disclosure via path traversal vulnerability
NLTK version 3.9.4 is vulnerable to a path traversal attack due to an incomplete fix for GitHub Issue #3504. The `_UNSAFE_NO_PROTOCOL_RE` regex in `nltk/data.py` checks for literal `../` sequences but fails to account for percent-encoded traversal sequences such as `..%2f`. The `url2pathname()` function decodes these sequences after the validation step, allowing an attacker to bypass the protection. This vulnerability enables an attacker to read arbitrary files accessible to the Python process by controlling the resource name parameter passed to `nltk.data.load()` or `nltk.data.find()`. The issue affects applications that rely on NLTK for resource loading, including NLP web applications, Jupyter notebooks, and CLI tools. The default `pathsec.ENFORCE=False` setting exacerbates the impact by not blocking the file read at the `open()` stage. A flaw was found in NLTK. An attacker can exploit a path traversal vulnerability by providing specially crafted input to `nltk.data.load()` or `nltk.data.find()`. The vulnerability allows arbitrary file reads when an attacker can control the resource_name parameter by using percent-encoded path separators to bypass validation. In Red Hat products where NLTK is bundled (OpenShift AI, OpenShift Lightspeed, Ansible Automation Platform), the resource loading functions are typically used internally to load pre-packaged NLP models and corpora.
High [CVE-2026-55607] @anthropic-ai/claude-code: Claude Code: Arbitrary code execution through git directory confusion
Claude Code is an agentic coding tool. From 2.1.38 until 2.1.163, Claude Code's worktree handling allowed creation of worktrees named ".git" and navigation to worktrees outside the sandbox context, enabling git directory confusion attacks. By exploiting symlink manipulation and git fsmonitor execution during worktree operations, an attacker could overwrite files in the user's home directory (such as.zshenv), leading to code execution outside of seatbelt sandbox restrictions. Reliably exploiting this required the user to clone a malicious repository containing prompt injection content and run Claude Code against it. This vulnerability is fixed in 2.1.163. This vulnerability allowed the creation of specially named worktrees and navigation outside of the intended secure environment, leading to what is known as a 'git directory confusion attack'. By manipulating symbolic links and how git monitors file system changes, an attacker could overwrite important user files, potentially leading to unauthorized code execution on the user's system. This attack requires a user to interact with a malicious code repository. Red Hat severity: Important — CVSS 7.1 (CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:U/C:H/I:H/A:H). Weakness: CWE-59. Affected Red Hat products: OpenShift Lightspeed. Red Hat does not currently list a fixing RHSA for this CVE.
High [CVE-2026-13676] Security policy bypass due to improper Unicode hostname canonicalization
fast-uri versions 2.3.1 through 3.1.2 and 4.0.0 fail to canonicalize Unicode (IDN) hostnames for HTTP-family URLs. The IDN conversion path calls a helper that does not exist on the global URL constructor, silently leaving the host in its original Unicode form while normalize() and equal() still return values that differ from a WHATWG-compatible URL parser. Applications that use fast-uri to enforce host-based policy (denylists, loopback filtering, redirect validation, outbound proxy routing) before passing the same URL to Node's URL or fetch can be bypassed when the two implementations resolve the same input to different hosts. Patches: upgrade to fast-uri 3.1.3 for the 3.x line or 4.0.1 for the 4.x line. Workarounds: enforce host policy using the same URL parser used for the actual request, or reject non-ASCII hosts before policy checks. A remote attacker could exploit this to circumvent security controls and potentially access unauthorized resources or perform malicious redirects. This Important flaw in `fast-uri` allows a remote attacker to bypass host-based security policies. Applications that rely on `fast-uri` for URL parsing and policy enforcement, such as denylists or redirect validations, can be circumvented due to inconsistent handling of Unicode hostnames.