CVE Vulnerability Catalog

Heap buffer overflow in PostgreSQL regexp allows the query author to execute arbitrary code as the operating system user running the database, via text that would not pass encoding validation. This shares heritage with CVE-2026-2006, but this case involved unanticipated data growth when round-tripped through pg_wchar. Versions before PostgreSQL 18.6, 17.11, 16.15, 15.19, and 14.24 are affected.

In PostgreSQL, the pgcrypto extension stores cleartext when ciphers are disabled, allowing a user to recover cleartext via direct observation of faulty ciphertext. The OpenSSL version and configuration determine disabled ciphers. If the application accepts encrypted data as input, decryption may succeed even with the wrong key, losing MDC protection. Affected versions are before 18.6, 17.11, 16.15, 15.19, and 14.24.

Integer wraparound in PostgreSQL tsvector and tsquery data type functions allows an unprivileged database user to cause the server to undersize an allocation and write out-of-bounds, via crafted large inputs. This may execute arbitrary code as the operating system user running the database. These types are typically sourced from application logic, not taken from the application's user. Hence, application users attacking the database, through the application as a conduit, are unlikely. CVE-2026-6473 had fixed similar problems. Versions before PostgreSQL 18.6, 17.11, 16.15, 15.19, and 14.24 are affected.

HCL AION is affected by a vulnerability where the shared storage used by product components is architected without sufficient access separation. Processes sharing the storage may be able to access or modify files beyond their intended scope, potentially resulting in unintended behavior or security impact under certain conditions.

Serendipity before 2.6.0 contains a server-side request forgery vulnerability in the serendipity_url_allowed() filter that fails to block hex-encoded IPv4 addresses, IPv6 literals, and link-local ranges. Authenticated users with adminImagesAdd permission can bypass the filter using alternate address formats to request internal services and retrieve response bodies through the public uploads directory.

Serendipity versions >= 2.3.5 and <= 2.6.0 contain a reflected cross-site scripting vulnerability in the search clean-URL route (/search/<term>). In include/functions_routing.inc.php serveSearch(), the sanitisation pipeline runs urldecode() after HTML-encoding, so a single URL-encoded HTML payload survives strip_tags() and htmlspecialchars() and is then decoded back into live HTML in the page. A crafted search link can execute arbitrary JavaScript in the victim's browser. Fixed in 2.6.1.

JupyterLab (pip package 'jupyterlab') versions >=4.1.0,<=4.5.9 and >=4.6.0,<=4.6.1 contain a plugin manager lock-rule enforcement bypass. Two server-side enforcement gaps allow an authenticated user to circumvent administrator lock rules by making direct requests to the /lab/api/plugins endpoint, enabling or disabling plugins that were locked — including child plugins of multi-plugin extensions and plugins locked via the 'lock all' mechanism. This can impact data integrity and bypass hardening or restrictions (e.g., download/upload limits) implemented through locked plugins. Fixed in versions 4.6.2 and 4.5.10.

JupyterLab versions >=4.6.0,<=4.6.1 and <=4.5.9 contain an allowlist/blocklist enforcement gap in PyPIExtensionManager.install(). A missing 'await' caused the is_install_allowed coroutine to never execute, so the extension allowlist/blocklist check was not enforced for direct callers of install(). The stock JupyterLab HTTP API and Extension Manager UI are not affected, as they perform a separate, correctly awaited check. The issue affects only deployments where a custom extension or downstream integration imports PyPIExtensionManager and calls install() directly with a package name influenced by untrusted input, an allowlist/blocklist is configured, the PyPI Extension Manager is enabled, and kernels and terminals are disabled or delegated to remote hosts. Fixed in JupyterLab 4.6.2 and 4.5.10.

GitPython versions before 3.1.54 contain a remote code execution vulnerability in the check_unsafe_options guard that can be bypassed by smuggling git options inside single-character kwarg values. Attackers can supply crafted option dictionaries to clone_from, fetch, pull, push, ls_remote, iter_commits, blame, or archive methods to execute arbitrary OS commands via the --upload-pack parameter.

GitPython versions before 3.1.54 contain an arbitrary file overwrite vulnerability in the Diffable.diff method that fails to validate git options passed through kwargs. Attackers can supply the --output argument via the other parameter or output kwarg to write patch content to attacker-chosen file paths at process privilege level.

GitPython before 3.1.54 contains an incomplete denylist in unsafe_git_clone_options that omits --template, allowing attackers to achieve arbitrary command execution during clone operations. Attackers can supply --template pointing to a directory containing malicious post-checkout hooks that execute when git clones the repository.

GitPython before 3.1.55 fails to disable environment variable expansion in Remote.create() and Submodule.add() URL handling, allowing attackers to exfiltrate secrets by supplying URLs containing variable references. Attackers can craft URLs with environment variable tokens that are expanded into .git/config and .gitmodules, then transmitted to attacker-controlled hosts during fetch or pull operations.

GitPython before 3.1.56 contains an argument injection vulnerability in the Commit.count() method, which forwards keyword arguments to 'git rev-list' without the check_unsafe_options guard present in the sibling iter_items method. An attacker who can control options passed to Commit.count (e.g., via an application that forwards a user-supplied options dict) can supply output=<path>, causing 'git rev-list --output=<path>' to open and truncate the target file to zero bytes before revision parsing. This allows destruction/blanking of an arbitrary file at the process's privilege level (no content control, 0-byte truncation).

GitPython before 3.1.57 fails to guard git option forwarding in IndexFile.checkout() and TagReference.create(), allowing attackers to pass unsafe options via kwargs. Attackers can use --prefix to overwrite arbitrary files with repository content or -F to read arbitrary files returned in-band.

GitPython before 3.1.57 contains an incomplete denylist in the unsafe_git_archive_options guard that omits --add-file and --add-virtual-file options. Attackers can supply these options to Repo.archive() to read arbitrary files from the filesystem and include them in the returned archive.

Budibase Server before 3.40.0 contains a NoSQL injection vulnerability in the MongoDB query execution endpoint where user-supplied parameters are interpolated into JSON query templates without proper sanitization of JSON metacharacters. Attackers with query write permission can inject JSON structural characters to alter MongoDB queries, bypassing filters to read, modify, or delete arbitrary documents.

Budibase before 3.40.0 contains a NoSQL injection vulnerability in the MongoDB datasource integration where user-supplied parameters are enriched with handlebars using noEscaping: true and parsed without operator filtering. Attackers can inject MongoDB operators through query parameters to bypass per-user access controls, read arbitrary documents, execute JavaScript via $where operators, or modify collections through update and delete operations.

OpenRemote notification deletion endpoints fail to enforce realm boundaries, allowing any realm administrator to delete notifications belonging to other realms. Attackers with write:admin role in one realm can send DELETE requests to remove notifications from the master realm or other tenants without authorization checks.

Network-AI before version 5.15.1 contains a security matcher bypass vulnerability where SandboxPolicy evaluates raw command strings with quotes preserved while the executor tokenizes commands by stripping quotes before execution. Attackers can craft quoted commands that evade blocklist checks and approval gates while the executor runs the identical unquoted dangerous argv.

Network-AI ClaudeHookBridge before 5.15.1 truncates the target string to 500 characters before evaluating denyPatterns, while Claude Code executes the full untruncated command. Attackers can position dangerous content past byte 500 in a Bash command field to bypass the operator's hard-deny list and execute arbitrary commands.