| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| ImageMagick before 7.1.2-15 and 6.9.x before 6.9.13-40 contains an integer overflow in the PSB (PSD v2) RLE decoding path (ReadPSDChannelRLE in coders/psd.c) that causes a heap out-of-bounds read on 32-bit builds. Processing a crafted PSB file can lead to information disclosure or a crash. |
| Backpropagate is a Python library for fine-tuning large language models on a single GPU. In versions 1.1.0 and 1.1.1, the optional Reflex web UI exposes a training control plane without authentication: dataset upload, model load, training start/stop, multi-run orchestration, GGUF export, and HuggingFace Hub push. The CLI accepts two operator-facing flags intended as security controls: --auth user:pass — documented as "require HTTP Basic authentication on every request to the UI." and--share — documented as "expose the UI on a public address; requires --auth." When --auth user:pass is passed, the CLI prints Auth: enabled (user: <username>) to confirm to the operator that authentication is active, then exports BACKPROPAGATE_UI_AUTH=user:pass to the subprocess that launches the Reflex backend. The Reflex backend (backpropagate/ui_app/**) never reads BACKPROPAGATE_UI_AUTH. No authentication middleware is registered. No request-level guard runs. No WebSocket upgrade guard runs. Any client that reaches the bound port — local or remote, depending on whether --share is used — has full UI access. An inline comment at backpropagate/cli.py:1217-1218 in the v1.1.0 source documents the gap: "For Phase 1 the variable is exported but Reflex doesn't read it yet." This comment was internal-facing; the user-facing documentation (README, CHANGELOG, SHIP_GATE) advertised the contract as enforced. An attacker who reaches the bound port can read uploaded datasets, trigger arbitrary training runs against any local base models as well as read their paths, trigger HuggingFace Hub pushes and cause disk-fill DoS. This issue has been fixed in version 1.2.0. If developers cannot immediately upgrade to 1.2.0 run backprop ui with no flags so it binds to localhost, use SSH port-forwarding (ssh -L 7860:localhost:7860 <training-host>) instead of --share for remote access, and audit any host previously launched with --share, re-issuing any HF tokens used during those sessions. |
| Unauthenticated Sensitive Data Exposure in JetBlog <= 2.4.8 versions. |
| The shell tool command allowlist in the SecurityPolicy of OpenHuman desktop agent through 0.54.0 (default Supervised security policy) can be bypassed to execute arbitrary OS commands with the privileges of the desktop user. Two flaws in src/openhuman/security/policy.rs combine: (1) is_args_safe() blocks the find flags -exec and -ok but not the functionally identical -execdir and -okdir, which also execute an arbitrary command for each matched file; and (2) skip_env_assignments() strips leading inline KEY=value environment-variable assignments before allowlist validation, so a command such as GIT_EXTERNAL_DIFF=<cmd> git diff is validated as the allowed git diff but, when executed via the shell, runs <cmd> through git's environment-driven hooks (for example GIT_EXTERNAL_DIFF or GIT_SSH_COMMAND). Because the sandbox is the primary trust boundary between untrusted LLM-processed content and the host operating system, an attacker can achieve remote code execution via indirect prompt injection: a malicious document, email, calendar event, or web page ingested by the agent instructs it to run a benign-looking allowlisted command, resulting in arbitrary command execution, data exfiltration, arbitrary file read/write, and lateral movement on the user's machine. The issue was fixed in commit 60050aa09a870f53ed7e4cd40ed41fd2860329e7 (first released in 0.54.22-staging; first stable release 0.56.0), which blocks -execdir/-okdir for find. |
| picklescan before 0.0.33 contains an incomplete deny-list that fails to block pydoc.locate and operator.methodcaller functions, allowing attackers to bypass security checks. Remote attackers can craft malicious pickle files using these unblocked functions to achieve arbitrary code execution when the pickle is deserialized. |
| picklescan before 0.0.33 fails to block the ctypes module, allowing attackers to achieve remote code execution by invoking direct syscalls and accessing raw memory. Attackers can craft malicious pickle files using ctypes.WinDLL to load kernel32.dll and execute arbitrary commands, bypassing sandbox protections and gadget chain detection. |
| picklescan before 1.0.4 contains an incomplete blocklist for the profile module that fails to block the module-level profile.run() function, allowing attackers to achieve arbitrary code execution via exec(). Attackers can craft malicious pickle files calling profile.run(statement) to execute arbitrary Python code while picklescan reports zero security issues. |
| Pimcore CMS/DXP version 12.3.8 contains a sandbox bypass vulnerability that allows authenticated administrative attackers to execute arbitrary methods on PHP objects by exploiting empty checkMethodAllowed() and checkPropertyAllowed() implementations in the custom Twig SecurityPolicy. Attackers can supply malicious Twig templates through the DataObject ClassDefinition Layout\Text component to perform arbitrary file reads, execute arbitrary database queries, and potentially achieve remote code execution via PHP object gadget chains, with the pimcore_* function wildcard further broadening the bypass to all Pimcore Twig functions. |
| 8cc is vulnerable to an Out‑of‑Bounds Read due to improper handling of #line directives and GNU linemarkers. The compiler accepts attacker-controlled filename and line number metadata and later uses it without validation when accessing source line arrays.
By supplying invalid or oversized line numbers, an attacker can trigger out-of-bounds memory access and a crash.
Maintainer of this project was notified early about this vulnerability, but didn't respond with the details of vulnerability or vulnerable version range. Version corresponding to the commit b480958 was tested and confirmed as vulnerable, other versions were not tested but might also be vulnerable. |
| NILFS utilities through 2.3.0, fixed in commit 26efb5d, nilfs_sb_is_valid() function fails to validate s_log_block_size field in NILFS2 superblock before bit-shift operations. Attackers supplying crafted NILFS2 images trigger undefined behavior through oversized shifts or out-of-memory conditions, crashing tools like nilfs-tune and dumpseg. |
| The deprecated functions ns_printrrf, ns_printrr and fp_nquery in the GNU C Library version 2.0.1 to version 2.43 fail to validate the RDATA content against the RDATA length in a DNS response when processing A6, CERT, LOC, TKEY or TSIG records, which may allow an attacker to craft a DNS response, causing a target application to crash or read uninitialized memory.
These functions are for application debugging only and hence not in the path of code executed by the DNS resolver. Further, they have been deprecated since version 2.34 and should not be used by any new applications. Applications should consider porting away from these interfaces since they may be removed in future versions. |
| Incorrect calculation of buffer size in Windows VMSwitch allows an authorized attacker to deny service locally. |
| Net::Statsd::Lite versions through 0.10.0 for Perl allowed metric injections.
The values from the set_add method were not checked for newlines, colons or pipes. Metrics generated from untrusted sources could inject additional statsd metrics.
Note that version 0.9.0 fixed a similar issue CVE-2026-46719 for metric names. |
| Net::Async::Statsd::Client versions through 0.005 for Perl allow metric injections.
The metric names are not checked for newlines, colons or pipes. Metrics generated from untrusted sources could inject additional statsd metrics. |
| Metrics::Any::Adapter::SignalFx versions before 0.04 for Perl does not protect against metric injections.
The statsd protocol (and extensions such as dogstatsd) allow mutiple metrics, separated by newlines, to be sent per packet.
Metrics::Any::Adapter::SignalFx which extends Metrics::Any::Adapter::Statsd, which has a similar vulnerability.
In addition, the _labels function does not check tags labels newlines or statsd control characters. The labels can be used for metric injections. |
| Metrics::Any::Adapter::DogStatsd versions before 0.04 for Perl does not protect against metric injections.
The statsd protocol (and extensions such as dogstatsd) allow mutiple metrics, separated by newlines, to be sent per packet.
Metrics::Any::Adapter::DogStatsd which extends Metrics::Any::Adapter::Statsd, which has a similar vulnerability.
In addition, the _tags function does not check tags for newlines or statsd control characters. The tags can be used for metric injections. |
| Metrics::Any::Adapter::Statsd versions before 0.04 for Perl does not protect against metric injections.
The statsd protocol (and extensions) allow mutiple metrics, separated by newlines, to be sent per packet.
The send method does not validate the contents of the metric names or values. If the names have newlines and statsd control characters (colon, pipe) then metric injections are possible.
Version 0.04 fixed this by modifying the _make method to block metric names with characters below ASCII 32 (which includes the newline), or colons or pipes. |
| Etsy::StatsD versions through 1.002002 for Perl allow metric injections.
The metric names and values are not checked for newlines, colons or pipes. Metrics generated from untrusted sources could inject additional statsd metrics.
Note that the git repository contains an unreleased version with the gauge and set methods that also do not check for potential metric injections. |
| Mojolicious::Plugin::Statsd versions through 0.04 for Perl allowed metric injections.
The metric names and set values were not checked for newlines, colons or pipes. Metrics generated from untrusted sources could inject additional statsd metrics.
Version 0.06 changes the module from being a statsd client to using a separate statsd client. It defaults to using a version of Net::Statsd::Tiny that fixes a similar issue (CVE-2026-46720). |
| Net::Statsd versions before 0.13 for Perl allow metric injections.
The metric names are not checked for newlines, colons or pipes. Metrics generated from untrusted sources could inject additional statsd metrics.
The update_stats (used for updating counters) and gauge methods do not check that values are numeric (which would block metric injection). |
