PT-2026-82206

Name of the Vulnerable Software and Affected Versions langfun versions prior to 0.1.2 Description An Eval Injection occurs in the default lf.query Python protocol. This issue allows remote unauthenticated attackers to execute arbitrary Python code within the host application context. The flaw is...

NeuronFuzz: Safety Neuron Guided Fuzzing for LLM Safety Evaluation

Safety evaluation is critical for assessing whether aligned Large Language Models LLMs remain robust against jailbreak attacks. Existing automated testing methods, however, largely rely on response-level feedback: each candidate prompt typically requires generating a target-model response to...

Improper Neutralization of Input Used for LLM Prompting

Overview strands-agents-tools is an A collection of specialized tools for Strands Agents Affected versions of this package are vulnerable to Improper Neutralization of Input Used for LLM Prompting via the pythonrepl function in src/strandstools/pythonrepl.py. An attacker can execute arbitrary...

CVE-2026-78379

Improper neutralization of input used for LLM prompting in the pythonrepl tool in Amazon Strands Agents Tools before 0.8.5 might allow remote actors to execute arbitrary Python code on the agent's host by bypassing the human consent gate, via a crafted prompt that forwards noninteractivemode as a...

EUVD-2026-65735

Improper neutralization of input used for LLM prompting in the pythonrepl tool in Amazon Strands Agents Tools before 0.8.5 might allow remote actors to execute arbitrary Python code on the agent's host by bypassing the human consent gate, via a crafted prompt that forwards noninteractivemode as a...

CVE-2026-55557

browse-mcp is a Playwright-based headless-browser MCP server for MCP-capable agents. Prior to 0.8.2 , the browser_download function writes a fetched response body to join(save_dir, filename) without validating the caller-controlled save_dir, and browser_save_state / browser_load_state honor a cal...

PraisonAI: Origin-validation bypass (startswith prefix match) enables unauthenticated cross-site request forgery against the PraisonAI MCP HTTP server

Summary The PraisonAI MCP server exposes an HTTP-stream transport praisonai mcp serve --transport http-stream that binds to localhost and, by default, has no API key. Its only access control for browser-originated requests is an Origin allowlist, which the code implements as required by the MCP...

Grok fooled into stealing user chat, location data, and more

A new type of prompt injection attack shows why giving AI assistants access to browsers, code tools, and private data deserves extra caution. AI researchers describe "Cryptographic Context Injection"—an attack that hides malicious instructions inside encrypted data. The AI is then persuaded to...

CVE-2026-45134

A flaw was found in the LangSmith Client SDKs. This vulnerability allows an external attacker to control the content of prompt manifests when a user pulls a public prompt from the LangSmith Hub. By exploiting the deserialization of untrusted data, an attacker could achieve high confidentiality...

Prompt Structure Redistributes, Not Reduces: An Empirical Analysis of Security-Weaknesses in LLM-Generated Python Code

Large Language Models LLMs increasingly generate code from natural-language prompts, making prompt engineering a key mechanism for shaping the security of generated software. Structured and security-oriented prompts are widely used to encourage safer code, yet their effects extend beyond whether...

PT-2026-81489

Name of the Vulnerable Software and Affected Versions Amazon Strands Agents Tools versions prior to 0.8.5 Description Improper neutralization of input used for Large Language Model LLM prompting in the python repl tool allows remote actors to execute arbitrary Python code on the agent's host. Thi...

Evaluating and Preventing Security Smells in AI-Generated Ansible Code

AI coding assistants generate Infrastructure as Code, yet no work has examined whether this code meets security requirements. This matters because security smells in infrastructure code propagate to deployed systems, producing infrastructure that is insecure and untrustworthy. We evaluate 16 AI...

CVE-2026-76072

The Continue CLI applies an incomplete denylist as its only barrier to destructive shell commands when running unattended. In headless mode and auto mode the default policy in extensions/cli/src/permissions/defaultPolicies.ts grants the Bash tool the allow permission, and permissionChecker.ts...

CVE-2026-76072 Continue CLI through 1.5.47 Incomplete Destructive Command Denylist in Headless and Auto Mode

The Continue CLI applies an incomplete denylist as its only barrier to destructive shell commands when running unattended. In headless mode and auto mode the default policy in extensions/cli/src/permissions/defaultPolicies.ts grants the Bash tool the allow permission, and permissionChecker.ts...

EUVD-2026-65059

The Continue CLI applies an incomplete denylist as its only barrier to destructive shell commands when running unattended. In headless mode and auto mode the default policy in extensions/cli/src/permissions/defaultPolicies.ts grants the Bash tool the allow permission, and permissionChecker.ts...

CVE-2026-76072

Continue CLI through 1.5.47 uses an incomplete denylist in evaluateTerminalCommandSecurity.ts as its sole barrier against destructive shell commands in headless and auto modes. The dangerous-path check covers only /, ~, /usr, /etc, /bin, and /sbin, leaving paths like /home, /root, /var, /opt, and...

CVE-2026-76072 Continue CLI through 1.5.47 Incomplete Destructive Command Denylist in Headless and Auto Mode

The Continue CLI applies an incomplete denylist as its only barrier to destructive shell commands when running unattended. In headless mode and auto mode the default policy in extensions/cli/src/permissions/defaultPolicies.ts grants the Bash tool the allow permission, and permissionChecker.ts...

CVE-2026-76072

The Continue CLI applies an incomplete denylist as its only barrier to destructive shell commands when running unattended. In headless mode and auto mode the default policy in extensions/cli/src/permissions/defaultPolicies.ts grants the Bash tool the allow permission, and permissionChecker.ts...

leash-poc

leash-poc: MCP tool-poisoning, contained by an independent pol...

hermetic

Der agentenisolierte Credential-Broker. Ihr KI-Agent führt den authentifizierten Aufruf durch. Er sieht nie den Schlüssel. -brightgreen?style=flat-square Webseite · Installation · Schnellstart · Funktionsweise · Sicherheitsmodell · Schwachstelle melden !IMPORTANT Ihr KI-Agent arbeitet auf der ★★★...