Incomplete List of Disallowed Inputs

Overview sagemaker-core is an An python package for sagemaker core functionalities Affected versions of this package are vulnerable to Incomplete List of Disallowed Inputs via the searchhub function. An attacker can execute arbitrary commands, access sensitive data, or compromise the execution...

CVE-2026-29612

OpenClaw versions prior to 2026.2.14 decode base64-backed media inputs into buffers before enforcing decoded-size budget limits, allowing attackers to trigger large memory allocations. Remote attackers can supply oversized base64 payloads to cause memory pressure and denial of service...

CVE-2026-29612 OpenClaw < 2026.2.14 - Denial of Service via Large Base64 Media File Decoding

OpenClaw versions prior to 2026.2.14 decode base64-backed media inputs into buffers before enforcing decoded-size budget limits, allowing attackers to trigger large memory allocations. Remote attackers can supply oversized base64 payloads to cause memory pressure and denial of service...

EUVD-2026-9936

OpenClaw versions prior to 2026.2.14 decode base64-backed media inputs into buffers before enforcing decoded-size budget limits, allowing attackers to trigger large memory allocations. Remote attackers can supply oversized base64 payloads to cause memory pressure and denial of service...

EUVD-2025-208315

An Arbitrary File Read vulnerability exists in the ImageTextPromptValue class in Exploding Gradients RAGAS v0.2.3 to v0.2.14. The vulnerability stems from improper validation and sanitization of URLs supplied in the retrievedcontexts parameter when handling multimodal inputs...

GHSA-V2XR-WVRV-P969 RAGAS has an Arbitrary File Read vulnerability

An Arbitrary File Read vulnerability exists in the ImageTextPromptValue class in Exploding Gradients RAGAS v0.2.3 to v0.2.14. The vulnerability stems from improper validation and sanitization of URLs supplied in the retrievedcontexts parameter when handling multimodal inputs...

RAGAS has an Arbitrary File Read vulnerability

An Arbitrary File Read vulnerability exists in the ImageTextPromptValue class in Exploding Gradients RAGAS v0.2.3 to v0.2.14. The vulnerability stems from improper validation and sanitization of URLs supplied in the retrievedcontexts parameter when handling multimodal inputs...

CVE-2025-45691

An Arbitrary File Read vulnerability exists in the ImageTextPromptValue class in Exploding Gradients RAGAS v0.2.3 to v0.2.14. The vulnerability stems from improper validation and sanitization of URLs supplied in the retrievedcontexts parameter when handling multimodal inputs...

CVE-2025-45691

An Arbitrary File Read vulnerability exists in the ImageTextPromptValue class in Exploding Gradients RAGAS v0.2.3 to v0.2.14. The vulnerability stems from improper validation and sanitization of URLs supplied in the retrievedcontexts parameter when handling multimodal inputs...

Malicious AI Assistant Extensions Harvest LLM Chat Histories

Microsoft Defender has been investigating reports of malicious Chromium‑based browser extensions that impersonate legitimate AI assistant tools to harvest LLM chat histories and browsing data. Reporting indicates these extensions have reached approximately 900,000 installs. Microsoft Defender...

Permissive List of Allowed Inputs

Overview @backstage/plugin-techdocs-node is a Common node.js functionalities for TechDocs, to be shared between techdocs-backend plugin and techdocs-cli Affected versions of this package are vulnerable to Permissive List of Allowed Inputs via the processing of the mkdocs.yml configuration file...

PT-2026-23468

Name of the Vulnerable Software and Affected Versions Exploding Gradients RAGAS versions 0.2.3 through 0.2.14 Description An arbitrary file read issue exists in the ImageTextPromptValue class. This is due to insufficient validation and sanitization of URLs provided in the retrieved contexts...

CVE-2025-45691

An Arbitrary File Read vulnerability exists in the ImageTextPromptValue class in Exploding Gradients RAGAS v0.2.3 to v0.2.14. The vulnerability stems from improper validation and sanitization of URLs supplied in the retrievedcontexts parameter when handling multimodal inputs...

CVE-2025-45691

An Arbitrary File Read vulnerability exists in the ImageTextPromptValue class in Exploding Gradients RAGAS v0.2.3 to v0.2.14. The vulnerability stems from improper validation and sanitization of URLs supplied in the retrievedcontexts parameter when handling multimodal inputs...

CVE-2025-45691

CVE-2025-45691 affects VibrantLabs RAGAS (up to v0.4.3); the vulnerability lies in improper validation of URLs in retrieved_contexts during multimodal input processing, enabling Server-Side Request Forgery (SSRF) and arbitrary file reads. Several connected sources describe exploitation via manipu...

GHSA-JJGJ-CPP9-CVPV OpenClaw Vulnerable to Local File Exfiltration via MCP Tool Result MEDIA: Directive Injection

Summary A malicious or compromised MCP Model Context Protocol tool server can exfiltrate arbitrary local files from the host system by injecting MEDIA: directives into tool result text content. OpenClaw's tool result processing pipeline extracts file paths from MEDIA: tokens without source-level...

AZL-79463 CVE-2026-27601 affecting package rsyslog 8.2308.0-5

Underscore.js is a utility-belt library for JavaScript. Prior to 1.13.8, the .flatten and .isEqual functions use recursion without a depth limit. Under very specific conditions, detailed below, an attacker could exploit this in a Denial of Service DoS attack by triggering a stack overflow...

AZL-79323 CVE-2026-27601 affecting package cyrus-sasl-bootstrap 2.1.28-4

Underscore.js is a utility-belt library for JavaScript. Prior to 1.13.8, the .flatten and .isEqual functions use recursion without a depth limit. Under very specific conditions, detailed below, an attacker could exploit this in a Denial of Service DoS attack by triggering a stack overflow...

Incomplete List of Disallowed Inputs

Overview openclaw is a 🦞 OpenClaw — Personal AI Assistant Affected versions of this package are vulnerable to Incomplete List of Disallowed Inputs via the --compress-program flag in the sort process when sort is manually added to the tools.exec.safeBins configuration. An attacker can execute...

Reliance on Untrusted Inputs in a Security Decision

Overview openclaw is a 🦞 OpenClaw — Personal AI Assistant Affected versions of this package are vulnerable to Reliance on Untrusted Inputs in a Security Decision via improper parsing of the X-Forwarded-For header behind trusted proxies. An attacker can influence security decisions tied to client...