CVE-2026-45312

RAGFlow (open-source RAG engine) is affected in 0.24.0 and earlier by a Jinja2 template injection in the prompt generator (rag/prompts/generator.py). This allows any authenticated user to execute arbitrary OS commands on the server via the SSTI chain, once a user registers and creates a Canvas wo...

PT-2026-44826

Name of the Vulnerable Software and Affected Versions RAGFlow versions prior to 0.24.1 Description A Server-Side Template Injection SSTI exists in the prompt generator located in rag/prompts/generator.py. This issue allows authenticated users to execute arbitrary operating system commands on the...

BadBone: Backdoor Attacks against Backbone Models in Visual Prompt Learning

Prompt learning is a new machine learning paradigm that has attracted ample attention due to its simplicity and proven efficacy. Despite its growing adoption, the security vulnerabilities associated with this paradigm remain underexplored. In this work, we take the first step to propose BadBone, ...

PT-2026-45054

Name of the Vulnerable Software and Affected Versions PraisonAI affected versions not specified Description The direct-prompt CLI automatically expands @url: mentions in raw prompt text before agent execution. The MentionsParser.process function handles these mentions by performing a direct HTTP...

How to Compare the Security of Code Written by Humans to LLM-Generated Code

Large language models LLMs are rapidly transforming how software is created and maintained. Comparing LLM-generated code against human-written standards is essential to determine whether these new tools uphold or erode the security baselines established by professional developers. Yet, we lack a...

From Prompt Injection to Persistent Control: Defending Agentic Harness against Trojan Backdoors

LLM agents are evolving from conversational chatbots to operational tools in real-world workspaces. In local agentic harnesses, an LLM can read and write files, call tools, and reuse workspace state across sessions. While such capabilities enhance utility, they also expose a new attack surface fo...

RAGFlow 安全漏洞

RAGFlow is an open-source RAG engine based on deep document understanding, developed by InfiniFlow. Versions of RAGFlow prior to 0.24.0 contain security vulnerabilities. These vulnerabilities stem from Jinja2 template injection in the prompt generator, which could allow any authenticated user to...

CVE-2026-45374 CodeWhale: task_create Insecure Defaults Enable RCE via Prompt Injection in Project Files

CodeWhale is a DeepSeek + MiMo coding agent in terminal. Prior to 0.8.26, the taskcreate tool spawns durable sub-agents that inherit two insecure defaults, allowshell defaults to true config.rs:1499: self.allowshell.unwraportrue and autoapprove defaults to true taskmanager.rs:297: autoapprove:...

CVE-2026-45374 CodeWhale: task_create Insecure Defaults Enable RCE via Prompt Injection in Project Files

CodeWhale is a DeepSeek + MiMo coding agent in terminal. Prior to 0.8.26, the taskcreate tool spawns durable sub-agents that inherit two insecure defaults, allowshell defaults to true config.rs:1499: self.allowshell.unwraportrue and autoapprove defaults to true taskmanager.rs:297: autoapprove:...

CVE-2026-45374

CVE-2026-45374 affects CodeWhale’s DeepSeek+MiMo task_create flow. Before version 0.8.26, sub-agents inherit two insecure defaults: allow_shell = true and auto_approve = true, enabling unrestricted, unapproved shell access after user approval of a task_create prompt. This can lead to remote comma...

CVE-2026-45374 CodeWhale: task_create Insecure Defaults Enable RCE via Prompt Injection in Project Files

CodeWhale is a DeepSeek + MiMo coding agent in terminal. Prior to 0.8.26, the taskcreate tool spawns durable sub-agents that inherit two insecure defaults, allowshell defaults to true config.rs:1499: self.allowshell.unwraportrue and autoapprove defaults to true taskmanager.rs:297: autoapprove:...

BIT-MLFLOW-2026-2614 Arbitrary File Read via Prompt Tag Source Validation Bypass in mlflow/mlflow

A vulnerability in the createmodelversion handler of mlflow/server/handlers.py in mlflow/mlflow versions 3.9.0 and earlier allows an unauthenticated remote attacker to read arbitrary files from the server's filesystem. The issue arises when a CreateModelVersion request includes the tag...

The Surface You Test Is Not the Surface That Breaks

Tool-augmented LLM agents are vulnerable to prompt injection: a third party who controls part of the agent's context can plant instructions that the agent then executes as if they came from the user. Current evaluations report a single attack success rate per model on one channel, the tool output...

Investigating Detection and Obfuscation of Prompt Injection Attacks against Software Reverse Engineering AI Agents

Agentic software reverse engineering systems are vulnerable to prompt injection attacks placed into the source code of executable binary files. This research demonstrates defensive tactics for detecting the presences of prompt injection strings in the decompiler output of adversarial example...

Strengthening Polymorphic Prompt Assembling: Dynamic Separator Generation against Emerging Prompt Injection Attacks

Polymorphic Prompt Assembling PPA defends LLM agents against prompt injections by randomly selecting separator pairs from a fixed pool to isolate user input from system instructions. Although effective, static pool reuse exposes a blast-radius vulnerability: once a separator leaks, it can be...

Minimal Prompt Perturbations Lead to Code Vulnerabilities: Prompt Fragility and Hidden-State Signals in Coding LLMs

LLM-based coding assistants are seeing rapid adoption, offering substantial gains in developer productivity. As organizations increasingly ship code these agents produce, the security of that code becomes critical. Prior work has shown that minor prompt perturbations degrade the functional...

Automatically Attacking Software Reverse Engineering AI Agents

Software tools for reverse engineering executable binary files, such as Ghidra, enable malware analysts to safely conduct robust static analysis without having access to original source code. Coupled with the analytic power of large language models LLM, agentic systems enabled with tools, such as...

poc-ccweb-unauth-rce

CVE — pqhaz3925/ccweb Unauthenticated RCE via Claude Code Cont...

CVE-2026-45134

LangSmith Client SDKs provide SDK's for interacting with the LangSmith platform. Prior to LangSmith SDK Python 0.8.0 and JS/TS 0.6.0, the LangSmith SDK's prompt pull methods pullprompt / pullpromptcommit in Python, pullPrompt / pullPromptCommit in JS/TS fetch and deserialize prompt manifests from...

CVE-2026-9366

A vulnerability was found in NousResearch hermes-agent 2026.4.23. The impacted element is the function scancontextcontent of the file agent/promptbuilder.py. The manipulation results in injection. The attack may be performed from remote. The exploit has been made public and could be used. The...