Code injection in `saved_model_cli` in TensorFlow

ImpactTensorFlow's savedmodelcli tool is vulnerable to a code injection:savedmodelcli run --inputexprs 'x=print"malicious code to run"' --dir ./--tagset serve --signaturedef servingdefaultThis can be used to open a reverse shell savedmodelcli run --inputexprs 'hello=exec"""\nimport...

PYSEC-2026-963 Code injection in `saved_model_cli` in TensorFlow

Impact TensorFlow's savedmodelcli tool is vulnerable to a code injection: savedmodelcli run --inputexprs 'x=print"malicious code to run"' --dir ./ --tagset serve --signaturedef servingdefault This can be used to open a reverse shell savedmodelcli run --inputexprs 'hello=exec"""\nimport...

erb: ERB: Arbitrary code execution via deserialization bypass

A flaw was found in ERB, a templating system for Ruby. An attacker who can trigger deserialization of untrusted data in a Ruby application can bypass existing protections. This vulnerability allows for arbitrary code execution by exploiting specific public methods that evaluate template source...

CVE-2026-14651 connorskees grass visitor denial of service

A vulnerability has been found in connorskees grass up to 0.13.4. The impacted element is the function grasscompiler::selector::extend/grasscompiler::evaluate::visitor. The manipulation leads to denial of service. The attack must be carried out locally. The exploit has been disclosed to the publi...

PoCHarness

...

Exploit for Eval Injection in Langflow

No d...

Agentic and Generative AI for Open-Source Intelligence and Cyber Investigations: Taxonomy, Evaluation, Challenges, and Future Directions

The rapid growth of publicly available digital information has rendered manual open-source intelligence OSINT analysis insufficient for modern intelligence, cybersecurity, and cyber investigation. Large language models LLMs and agentic AI systems, capable of tool use, multi-step reasoning, and...

EUVD-2026-36318

OpenClaw's browser act interactions could bypass private-network navigation checks...

OpenClaw's browser act interactions could bypass private-network navigation checks

Summary OpenClaw's browser control SSRF checks blocked direct navigation to private or loopback URLs, but some Playwright act interactions could trigger navigation after the initial check. A later browser evaluation could then read from the page reached by that action-triggered navigation. This...

Knowledge over Parameters: Evolving Smart Contract Vulnerability Detection

Smart contract vulnerabilities are predominantly logic bugs whose detection requires structured, step-by-step procedural knowledge of attack patterns and contract semantics. Existing LLM-based methods struggle to generate this knowledge automatically: prompt-based methods rely on manually crafted...

Malicious code in zyncmap (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 3a65a1106fa2bab6eb0b5982b289665b4b96a6ad86769a867f6e62fb73663f77 [email protected] advertises itself as an SVG sanitization/minification utility, but index.js exports an undocumented function getPlugin that, when...

dynaconf: jinja2: Dynaconf: Arbitrary code execution via Server-Side Template Injection

A flaw was found in dynaconf, a Python configuration management tool. This Server-Side Template Injection SSTI vulnerability occurs due to unsafe template evaluation in the @Jinja resolver when the jinja2 package is installed. A remote attacker could exploit this by embedding malicious template...

LIB-TRAP: Standard Cell Library Hardware Trojan Risk Assessment and Prevention

Vulnerabilities inherent to the fabless semiconductor manufacturing model have significantly increased the risk of malicious Hardware Trojan HT insertion, posing severe threats to hardware security. Several HT mitigation and detection strategies have been developed, and existing works explore the...

Modern Enterprises: How to Evaluate the Security and Compliance of Office Software

Learn how modern businesses can judge office software for ISO 27001 certification, GDPR-aligned data handling, encryption, and safer PDF workflows with clarity...

EnclaveX: End-To-End Confidential AI with CPU/GPU TEEs

Large Language Models LLMs have rapidly proliferated, driving widespread adoption of AI applications. Most deployments rely on centralized infrastructures such as Microsoft Azure, Google Cloud, or AWS, requiring users to share sensitive data and training or fine-tuning code. This dependence raise...

Exploit for Improper Control of Dynamically-Managed Code Resources in Kidocode Crawl4Ai

CVE-2026-53753 — Crawl4AI Unauthenticated Remote Code Executio...

PYSEC-2026-383 Langroid has WAF Bypass Leading to RCE in TableChatAgent

Affected Scope langroid = 0.59.31 Vulnerability Description CVE-2025-46724 fix bypass: TableChatAgent can call pandaseval tool to evaluate the expression. There is a WAF in langroid/utils/pandasutils.py introduced to block code injection CVE-2025-46724. However it can be bypassed due to literalok...

Arbitrary Code Execution in Pillow

Pillow through 10.1.0 allows PIL.ImageMath.eval Arbitrary Code Execution via the environment parameter, a different vulnerability than CVE-2022-22817 which was about the expression parameter...

llama-index-core Prompt Injection vulnerability leading to Arbitrary Code Execution

A vulnerability was identified in the executils class of the llamaindex package, specifically within the safeeval function, allowing for prompt injection leading to arbitrary code execution. This issue arises due to insufficient validation of input, which can be exploited to bypass method...

PYSEC-2026-397 llama-index-core Prompt Injection vulnerability leading to Arbitrary Code Execution

A vulnerability was identified in the executils class of the llamaindex package, specifically within the safeeval function, allowing for prompt injection leading to arbitrary code execution. This issue arises due to insufficient validation of input, which can be exploited to bypass method...