CVE-2026-6013

A vulnerability was detected in D-Link DIR-513 1.10. This vulnerability affects the function formSetRoute of the file /goform/formSetRoute of the component POST Request Handler. The manipulation of the argument curTime results in buffer overflow. The attack may be performed from remote. The explo...

MINI-68RC-F2QG-WV93

Bulletin has no description...

PT-2026-31954

OpenClaw before 2026.3.24 contains an authorization bypass vulnerability in the HTTP /v1/models endpoint that fails to enforce operator read scope requirements. Attackers with only operator.approvals scope can enumerate gateway model metadata through the HTTP compatibility route, bypassing the...

OpenClaw security vulnerabilities

OpenClaw is an open-source intelligent artificial assistant developed by OpenClaw. Versions of OpenClaw prior to 2026.3.24 contained security vulnerabilities. These vulnerabilities stemmed from authorization bypasses in the HTTP/v1/models endpoints, which failed to enforce the requirement for...

ADAM: A Systematic Data Extraction Attack on Agent Memory Via Adaptive Querying

Large Language Model LLM agents have achieved rapid adoption and demonstrated remarkable capabilities across a wide range of applications. To improve reasoning and task execution, modern LLM agents would incorporate memory modules or retrieval-augmented generation RAG mechanisms, enabling them to...

RRC Steganography

This is a proof of concept tool called Rotation Range-Coding RRC Steganography - an efficient and provably secure linguistic steganographic method that embeds secret messages into natural-language text generated by large language models. Included is the whitepaper discussing this tool called...

BadSkill: Backdoor Attacks on Agent Skills Via Model-In-Skill Poisoning

Agent ecosystems increasingly rely on installable skills to extend functionality, and some skills bundle learned model artifacts as part of their execution logic. This creates a supply-chain risk that is not captured by prompt injection or ordinary plugin misuse: a third-party skill may appear...

CLIP-Inspector: Model-Level Backdoor Detection for Prompt-Tuned CLIP Via OOD Trigger Inversion

Organisations with limited data and computational resources increasingly outsource model training to Machine Learning as a Service MLaaS providers, who adapt vision-language models VLMs such as CLIP to downstream tasks via prompt tuning rather than training from scratch. This semi-honest setting...

CVE-2026-40212

OpenStack Skyline before 5.0.1, 6.0.0, and 7.0.0 has a DOM-based Cross-Site Scripting XSS vulnerability in the console because document.write is used unsafely, which is relevant in scenarios where administrators use the console web interface to view instance console logs...

CVE-2026-40212

OpenStack Skyline contains a DOM-based XSS in the console interface prior to 5.0.1, 6.0.0, and 7.0.0 due to unsafe use of document.write when administrators view instance console logs. Root cause is unsafe DOM manipulation in the console web UI. Impact is cross-site scripting in the admin console...

CVE-2026-35619: Incorrect Authorization

OpenClaw before 2026.3.24 contains an authorization bypass vulnerability in the HTTP /v1/models endpoint that fails to enforce operator read scope requirements. Attackers with only operator.approvals scope can enumerate gateway model metadata through the HTTP compatibility route, bypassing the...

Rembg security vulnerabilities

Rembg is a tool developed by Daniel Gatis for deleting image backgrounds. Versions of Rembg prior to 2.0.75 contained security vulnerabilities, which were caused by insufficient validation of the HTTP server’s modelpath parameter, potentially allowing path traversal attacks...

CVE-2026-5982 D-Link DIR-605L POST Request formAdvNetwork buffer overflow

A vulnerability was found in D-Link DIR-605L 2.13B01. This vulnerability affects the function formAdvNetwork of the file /goform/formAdvNetwork of the component POST Request Handler. Performing a manipulation of the argument curTime results in buffer overflow. Remote exploitation of the attack is...

CVE-2026-40113

PraisonAI is a multi-agent teams system. Prior to 4.5.128, deploy.py constructs a single comma-delimited string for the gcloud run deploy --set-env-vars argument by directly interpolating openaimodel, openaikey, and openaibase without validating that these values do not contain commas. gcloud use...

CVE-2026-40113 PraisonAI has an Argument Injection into Cloud Run Environment Variables via Unsanitized Comma in gcloud --set-env-vars

PraisonAI is a multi-agent teams system. Prior to 4.5.128, deploy.py constructs a single comma-delimited string for the gcloud run deploy --set-env-vars argument by directly interpolating openaimodel, openaikey, and openaibase without validating that these values do not contain commas. gcloud use...

Node.js: Node.js: Information disclosure due to `fs.realpathSync.native()` bypassing filesystem read restrictions

A flaw was found in Node.js. The Node.js Permission Model, intended to restrict filesystem access, does not properly enforce read permission checks for the fs.realpathSync.native function. This vulnerability allows code operating under --permission with restricted --allow-fs-read flags to bypass...

Wasmtime has a possible panic when lifting `flags` component value

Impact Wasmtime contains a possible panic which can happen when a flags-typed component model value is lifted with the Val type. If bits are set outside of the set of flags the component model specifies that these bits should be ignored but Wasmtime will panic when this value is lifted. This pani...

EUVD-2026-21022

Wasmtime has a possible panic when lifting flags component value...

EUVD-2026-20990

Wasmtime: Panic when transcoding misaligned utf-16 strings...

Wasmtime: Panic when transcoding misaligned utf-16 strings

Impact Wasmtime's implementation of transcoding strings into the Component Model's utf16 or latin1+utf16 encodings improperly verified the alignment of reallocated strings. This meant that unaligned pointers could be passed to the host for transcoding which would trigger a host panic. This panic ...