CVE-2026-65698

The CVE-2026-65698 entry describes a path traversal vulnerability in Void 1.3.4’s AI agent file-reading tools. The read_file, ls_dir, get_dir_tree, and search_* tools allow network-adjacent attackers to inject instructions and read arbitrary host files outside the open workspace, bypassing worksp...

CVE-2026-65698 Void 1.3.4 Path Traversal via AI Agent File-Reading Tools

Void through 1.3.4 contains a path traversal vulnerability in the AI agent file-reading tools that allows network-adjacent attackers to read arbitrary host files outside the open workspace by injecting instructions into content the agent processes. Attackers can supply absolute paths or file://...

PT-2026-60979

A vulnerability was identified in nearai ironclaw up to 0.29.1. The affected element is the function validate path of the file src/tools/builtin/path utils.rs of the component write file. The manipulation leads to link following. Local access is required to approach this attack. The exploit is...

PYSEC-2026-2578 Langroid: Path traversal in the file tools allows read/write outside configured current directory

Summary Langroid's ReadFileTool and WriteFileTool appear to treat currdir as the intended working-directory boundary for file operations. However, the tools only change the process working directory to currdir and then operate on the user-supplied filepath without resolving and enforcing that the...

CVE-2026-50181

Langroid is a framework for building large-language-model-powered applications. Prior to version 0.64.0, Langroid's ReadFileTool and WriteFileTool appear to treat currdir as the intended working-directory boundary for file operations. However, the tools only change the process working directory t...

CVE-2026-50181 Langroid: Path traversal in the file tools allows read/write outside configured current directory

Langroid is a framework for building large-language-model-powered applications. Prior to version 0.64.0, Langroid's ReadFileTool and WriteFileTool appear to treat currdir as the intended working-directory boundary for file operations. However, the tools only change the process working directory t...

CVE-2026-50181 Langroid: Path traversal in the file tools allows read/write outside configured current directory

Langroid is a framework for building large-language-model-powered applications. Prior to version 0.64.0, Langroid's ReadFileTool and WriteFileTool appear to treat currdir as the intended working-directory boundary for file operations. However, the tools only change the process working directory t...

Langroid: Path traversal in the file tools allows read/write outside configured current directory

Summary Langroid's ReadFileTool and WriteFileTool appear to treat currdir as the intended working-directory boundary for file operations. However, the tools only change the process working directory to currdir and then operate on the user-supplied filepath without resolving and enforcing that the...

Relative Path Traversal

Overview langroid is a Harness LLMs with Multi-Agent Programming Affected versions of this package are vulnerable to Relative Path Traversal through improper validation of user-supplied file paths in the ReadFileTool and WriteFileTool components. An attacker can access or modify files outside the...

PT-2026-55462

Name of the Vulnerable Software and Affected Versions Langroid versions prior to 0.64.0 Description Langroid's ReadFileTool and WriteFileTool fail to properly enforce the working-directory boundary defined by curr dir. While the tools change the process working directory to curr dir, they do not...

PYSEC-2026-303 CAI find_file Agent Tool has Command Injection Vulnerability Through Argument Injection

Summary The CAI Cybersecurity AI framework contains multiple argument injection vulnerabilities in its function tools. User-controlled input is passed directly to shell commands via subprocess.Popen with shell=True, allowing attackers to execute arbitrary commands on the host system. Vulnerable...

CVE-2026-54012

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.6, Open WebUI lets a user who can create, update, or import workspace models store arbitrary meta.knowledge entries on their model without checking whether they own or can read the...

PraisonAI: Compute-bridged file tools allow shell command injection

Compute-bridged file tools allow shell command injection Summary LocalManagedAgent / SandboxedAgent compute bridging wraps readfile, listfiles, and writefile when a compute provider is attached. The bridge converts those file operations into shell command strings using raw path arguments, then...

CVE-2026-6620

A vulnerability was found in SonicCloudOrg sonic-server up to 2.0.0. The affected element is the function Upload of the file FileTool.java of the component File Upload Endpoint. The manipulation of the argument Type results in path traversal. The attack may be launched remotely. The exploit has...

EUVD-2026-31565

A security flaw has been discovered in NousResearch hermes-agent up to 2026.4.16. This vulnerability affects the function isblockeddevice of the file tools/filetools.py of the component readfile Tool. Performing a manipulation results in path traversal. The attack may be initiated remotely. The...

CVE-2026-9351

A security flaw has been discovered in NousResearch hermes-agent up to 2026.4.16. This vulnerability affects the function isblockeddevice of the file tools/filetools.py of the component readfile Tool. Performing a manipulation results in path traversal. The attack may be initiated remotely. The...

Hermes Agent 路径遍历漏洞

Hermes Agent is an AI agent tool developed by Nous Research, featuring self-learning capabilities. Versions of Hermes Agent prior to 2026.4.16 contained a path traversal vulnerability. This vulnerability originated from the isblockeddevice function in the tools/filetools.py file within the readfi...

PT-2026-42907

Name of the Vulnerable Software and Affected Versions NousResearch hermes-agent versions prior to 2026.4.17 Description A flaw in the read file Tool within the tools/file tools.py file affects the is blocked device function. This issue allows a remote attacker to perform a path traversal, which i...

Cybersecurity AI (CAI) Framework 0.5.10 - Command Injection

Exploit Title: Cybersecurity AI CAI Framework 0.5.10 - Command Injection CVE: CVE-2026-25130 Date: 2026-02-03 Exploit Author: Mohammed Idrees Banyamer Author Country: Jordan Instagram: @banyamersecurity Author GitHub: https://github.com/yourusername Vendor Homepage:...

CVE-2026-7400 geekgod382 filesystem-mcp-server read_file_tool/write_file_tool server.py is_path_allowed path traversal

A security vulnerability has been detected in geekgod382 filesystem-mcp-server 1.0.0. This issue affects the function ispathallowed of the file server.py of the component readfiletool/writefiletool. Such manipulation leads to path traversal. The attack can be launched remotely. The exploit has be...