A vulnerability in the Linux kernel's xen/gntdev ioctl error handling was fixed. The flaw caused a memory leak and dangling list entry when gntdev_ioctl_map_grant_ref() failed to copy results back to userspace after adding a mapping. The issue stemmed from not releasing a reference acquired by gntdev_alloc_map() on error. Additionally, the mapping index returned to userspace could be incorrect due to modifications by gntdev_add_map(). The fix involves proper mutex handling, retrieving the correct index, and cleanup on copy_to_user() failure.
GitPython contains unguarded forwarding of git command options in two methods, IndexFile.checkout() and TagReference.create(), allowing arbitrary file overwrite and arbitrary file read respectively. The checkout method forwards unsafe options to git checkout-index, enabling an attacker to overwrite files anywhere on the filesystem with repository-controlled content via the --prefix option. The TagReference.create method forwards unsafe options to git tag, allowing reading of arbitrary files via the -F/--file option, returning their contents in-band. Both issues are present in the latest release and HEAD as of the report date. A fix is available in version 6.6_5.
Russh is a Rust SSH client & server library. Prior to 0.62.4, an authenticated SSH client can cause a denial of service by sending a pty-req channel request with more than 130 terminal-mode records. The parser in russh/src/server/encrypted.rs stores terminal modes in a fixed 130-entry [(Pty::TTY_OP_END, 0); 130] array but continues increasing the mode count, then constructs an out-of-bounds slice and panics before the application pty_request handler runs. The panic terminates the server session task without causing memory corruption. This issue is fixed in version 0.62.4.
CVE-2026-19752 is a server-side request forgery (SSRF) vulnerability in the EnzoVezzaro mcp-dominican-layer project affecting the parse-pdf function in src/index.ts. The vulnerability allows remote attackers to manipulate the pdfUrl argument to induce the server to make unauthorized requests. The issue affects versions up to commit 39dd373786712650097ad31db27d5c477c8f9c82. The vulnerability has a medium severity rating with a CVSS 4.0 score of 5.3. Although an exploit has been publicly disclosed, the project has not yet responded or issued a patch.
Previously, resolving relative paths containing parent directory ('..') segments performed string conversions and buffer rewrites on each step, resulting in quadratic time complexity and high memory allocation overhead. Now, path resolution operates on a byte buffer using index-based backtracking for '..' segments, eliminating the quadratic time complexity and significantly reducing memory allocations.
A flaw has been found in EnzoVezzaro mcp-dominican-layer up to 39dd373786712650097ad31db27d5c477c8f9c82. The affected element is the function axios.get of the file src/index.ts of the component parse-csv tool. This manipulation of the argument csvUrl causes server-side request forgery. The attack is possible to be carried out remotely. The exploit has been published and may be used. This product adopts a rolling release strategy to maintain continuous delivery. Therefore, version details for affected or updated releases cannot be specified. The project was informed of the problem early through an issue report but has not responded yet.
Improper Privilege Management (CWE-269) in Kibana Fleet can lead to privilege escalation via Privilege Escalation (CAPEC-233). An integration policy may optionally declare extra data streams that the integration writes to, which Fleet adds to the Elasticsearch API key issued to Elastic Agents enrolled in the corresponding agent policy. The resulting key allows new documents to be inserted and index mappings to be extended for specific indices. The key does not allow reading, updating, or deleting existing documents
Elasticsearch does not enforce an upper bound on a user-supplied count accepted by a search highlighting option, and the allocation derived from that count is not accounted against any circuit breaker. An authenticated user holding only read privileges on a single searchable index can submit one small search request that causes the node to reserve an excessively large internal data structure. The allocation occurs before the existing highlighting safety limits are evaluated, so memory exhaustion raises a fatal error that terminates the Elasticsearch node process. This results in a denial of service for the affected node and degrades cluster routing and health. The defect is not volumetric and does not depend on the size of the indexed data, so a single request is sufficient.
Elasticsearch does not apply its configurable input length restriction to a user-supplied pattern accepted by an intervals query. Compiling a deeply nested pattern drives unbounded recursion that exhausts the thread stack and raises a fatal error, terminating the Elasticsearch node process and causing a denial of service for that node. An authenticated user holding only read-only privileges on a single searchable index can trigger the condition with one small search request.
A flaw in Elasticsearch allows a low-privileged authenticated user who can index documents to submit a single small document containing a crafted user-supplied input. Processing one such document occupies a worker thread from a bounded pool for a disproportionate amount of time, degrading the availability of indexing operations on the affected node.