9 matches found
reverse-captcha-eval
リバースCAPTCHA: 大規模言語モデルの不可視ユニコード命令インジェクションに対する脆弱性評価 通常のテキストに埋め込まれた不可視のUnicodeエンコード命令を大規模言語モデルが実行するかどうかをテストする評価フレームワーク。従来のCAPTCHAが人間には解けるが機械には解けないタスクを利用するのに対し、リバースCAPTCHAは機械がアクセスできるが人間にはできない知覚チャネルを利用します。 論文: paper/reversecaptcha.pdf 主な発見 5つのモデル（GPT-5.2、GPT-4o-mini、Claude Opus 4、Claude Sonnet 4、Claud...
System-Velocity
System-Velocity Proxy-level security and maintenance for Veloci...
DEBIAN-CVE-2017-7481
Ansible before versions 2.3.1.0 and 2.4.0.0 fails to properly mark lookup-plugin results as unsafe. If an attacker could control the results of lookup calls, they could inject Unicode strings to be parsed by the jinja2 templating system, resulting in code execution. By default, the jinja2...
ansible: Security issue with lookup return not tainting the jinja2 environment
An input validation flaw was found in Ansible, where it fails to properly mark lookup-plugin results as unsafe. If an attacker could control the results of lookup calls, they could inject Unicode strings to be parsed by the jinja2 templating system, resulting in code execution. By default, the...
ansible: Security issue with lookup return not tainting the jinja2 environment
An input validation flaw was found in Ansible, where it fails to properly mark lookup-plugin results as unsafe. If an attacker could control the results of lookup calls, they could inject Unicode strings to be parsed by the jinja2 templating system, resulting in code execution. By default, the...
ansible: Security issue with lookup return not tainting the jinja2 environment
An input validation flaw was found in Ansible, where it fails to properly mark lookup-plugin results as unsafe. If an attacker could control the results of lookup calls, they could inject Unicode strings to be parsed by the jinja2 templating system, resulting in code execution. By default, the...
ansible: Security issue with lookup return not tainting the jinja2 environment
An input validation flaw was found in Ansible, where it fails to properly mark lookup-plugin results as unsafe. If an attacker could control the results of lookup calls, they could inject Unicode strings to be parsed by the jinja2 templating system, resulting in code execution. By default, the...
ansible: Security issue with lookup return not tainting the jinja2 environment
An input validation flaw was found in Ansible, where it fails to properly mark lookup-plugin results as unsafe. If an attacker could control the results of lookup calls, they could inject Unicode strings to be parsed by the jinja2 templating system, resulting in code execution. By default, the...
Microsoft Visual Basic 6.0 Project (Description) Stack overflow PoC
Exploit for unknown platform in category dos / poc =================================================================== Microsoft Visual Basic 6.0 Project Description Stack overflow PoC =================================================================== !/usr/local/bin/perl Discovered By UmZ Umair...