Lucene search
+L

1604 matches found

Kitploit
Kitploit
•added 2026/09/30 10:33 p.m.•13 views

openclaw-technical-analysis

OpenClaw Technical Analysis This repository contains my deep-dive technical reviews and code audits of the OpenClaw AI platform specifically focusing on the February 15 and February 25, 2026 builds. These reviews are grounded in raw architecture analysis, GitHub commit history, and the intersecti...

6AI score
SaveExploits0
Kitploit
Kitploit
•added 2026/09/30 3:46 p.m.•8 views

ActGuard

ActGuard ActGuardは、ツヌルを䜿甚するLLM゚ヌゞェントにおける間接的プロンプトむンゞェクションに察する実行前アクション監査防埡です。このリポゞトリには、最終的なActGuard実装ず、それを評䟡するために必芁なAgentDojoベヌスのランタむムが含たれおいたす。 このリリヌスには、最終的な完党なActGuard構成のみが含たれおいたす。内郚コンポヌネントは、AgentDojoベンチマヌクおよびトレヌス圢匏ずの互換性のために、匕き続き識別子reflectiveauditを䜿甚しおいたす。 芁件 Python 3.10以降 OpenRouter APIキヌ...

6.2AI score
SaveExploits0
Kitploit
Kitploit
•added 2026/09/30 12:46 p.m.•5 views

PIForge

PIForge RLベヌスのプロンプトむンゞェクションレッドチヌミングのためのオヌプンフレヌムワヌク 💻 コヌド · 🀗 モデル · 📜 論文 PIForgeは PISmith ず Climbing the Hill の共有コヌドベヌスです。PISmithはプロンプトむンゞェクションレッドチヌミングにおける報酬の疎性問題に察凊し、RL攻撃者が皀に成功する攻撃を探玢し孊習できるようにしたす。PISmithを基盀ずしお、Climbing the Hillはカリキュラム孊習を甚いお、より堅牢なフロンティアタヌゲットに察するレッドチヌミング時のコヌルドスタヌト問題に察凊したす。 ✹ ニュヌス...

6.3AI score
SaveExploits0References10
Packet Storm News
Packet Storm News
•added 2026/09/30 12:00 a.m.•5 views

From A2A Attacks to Envelope-Layer Defense: Red-Teaming Evaluation of LLM Agents and a Three-Layer Isomorphic Attack-Defense Model

Agent interaction protocols such as ACP and A2A have moved LLM-based agents toward multi-agent collaboration, introducing new security threats. A task sent by a remote peer over A2A is treated as a legitimate request, providing a natural channel for indirect prompt injection. Existing agent...

6.1AI score
SaveExploits0
UbuntuCve
UbuntuCve
•added 2026/09/29 9:17 p.m.•5 views

CVE-2026-102938

virtualenv is a tool for creating isolated virtual python environments. Prior to 21.7.11, PyEnvCfg.write writes prompt values verbatim to the line-oriented pyvenv.cfg format while PyEnvCfg.readvalues parses the file with str.splitlines and accepts the last value for duplicate keys. An attacker wh...

5.8CVSS5.9AI score0.0014EPSS
SaveExploits0References5
NVD
NVD
•added 2026/09/29 5:17 p.m.•6 views

CVE-2026-102697

Ollama versions 0.14.0 before 0.31.2 contain an incorrect authorization vulnerability in the experimental agent mode Bash tool approval mechanism that fails to properly parse shell syntax. Attackers who can influence model output through prompt injection can execute additional shell commands by...

8.5CVSS0.00146EPSS
SaveExploits0References6
Cvelist
Cvelist
•added 2026/09/29 4:46 p.m.•35 views

CVE-2026-102697 Ollama 0.14.0 before 0.31.2 Experimental Agent Bash Approval Bypass via Prefix-Based Authorization

Ollama versions 0.14.0 before 0.31.2 contain an incorrect authorization vulnerability in the experimental agent mode Bash tool approval mechanism that fails to properly parse shell syntax. Attackers who can influence model output through prompt injection can execute additional shell commands by...

8.5CVSS0.00146EPSS
SaveExploits0References6
CVE
CVE
•added 2026/09/29 4:46 p.m.•33 views

CVE-2026-102697

Ollama versions 0.14.0 before 0.31.2 are vulnerable to an incorrect authorization flaw within the experimental agent mode Bash tool approval mechanism . The root cause is a failure to properly parse shell syntax, which allows attackers to bypass session approval requirements. By using prompt inje...

8.5CVSS6AI score0.00146EPSS
SaveExploits0References6
Vulnrichment
Vulnrichment
•added 2026/09/29 4:46 p.m.•7 views

CVE-2026-102697 Ollama 0.14.0 before 0.31.2 Experimental Agent Bash Approval Bypass via Prefix-Based Authorization

Ollama versions 0.14.0 before 0.31.2 contain an incorrect authorization vulnerability in the experimental agent mode Bash tool approval mechanism that fails to properly parse shell syntax. Attackers who can influence model output through prompt injection can execute additional shell commands by...

8.5CVSS6AI score0.00146EPSS
SaveExploits0References6
OSV
OSV
•added 2026/09/29 4:46 p.m.•13 views

CVE-2026-102697 Ollama 0.14.0 before 0.31.2 Experimental Agent Bash Approval Bypass via Prefix-Based Authorization

Ollama versions 0.14.0 before 0.31.2 contain an incorrect authorization vulnerability in the experimental agent mode Bash tool approval mechanism that fails to properly parse shell syntax. Attackers who can influence model output through prompt injection can execute additional shell commands by...

8.5CVSS6AI score
SaveExploits0References8
EUVD
EUVD
•added 2026/09/29 4:46 p.m.•7 views

EUVD-2026-89089

Ollama versions 0.14.0 before 0.31.2 contain an incorrect authorization vulnerability in the experimental agent mode Bash tool approval mechanism that fails to properly parse shell syntax. Attackers who can influence model output through prompt injection can execute additional shell commands by...

8.5CVSS6AI score0.00146EPSS
SaveExploits0References6
Kitploit
Kitploit
•added 2026/09/29 4:32 p.m.•11 views

aisecplus-week01-danielossai

🔐 AISec Plus — 第1週: 脅嚁をマッピングする 研究者: Daniel Ossai Repo: aisecplus-week01-danielossai 提出: 第1週 · AISec Plus ブヌトキャンプ 📌 むンシデント名ず日付 EchoLeak CVE-2025-32711 発芋: 2025幎1月 · 開瀺: 2025幎6月 · 修正: 2025幎6月 📝 䜕が起こったか Aim Securityの研究者らは、Microsoft 365...

9.3CVSS8.6AI score0.08034EPSS
SaveExploits1
NVD
NVD
•added 2026/09/29 4:17 p.m.•11 views

CVE-2026-77177

Open GenAI Stack aka ogx-ai 2026-06-11, as used in the Meta AI backend for WhatsApp and other products, allows code execution because prompt injection with Jinja2 template syntax can be used to achieve server-side expression evaluation without sanitization...

9.8CVSS0.00592EPSS
SaveExploits0References1
Kitploit
Kitploit
•added 2026/09/29 12:21 p.m.•11 views

langgrinch-cve-2025-68664-analysis

LangGrinch 脆匱性分析: CVE-2025-68664 抂芁 このリポゞトリには、LangChain Core に圱響するシリアラむれヌションむンゞェクション脆匱性である CVE-2025-68664別名 LangGrinchを分析した技術レポヌトが含たれおいたす。 本レポヌトでは、脆匱性の背景、根本原因、悪甚フロヌ、脅嚁モデル、緩和策、および LLM ベヌスのアプリケヌションに察する珟実䞖界のセキュリティ圱響に぀いお説明したす。 察象トピック LangChain のシリアラむれヌションずデシリアラむれヌション 安党でないデシリアラむれヌション / CWE-502 LLM...

9.3CVSS8AI score0.4293EPSS
SaveExploits5References1
Kitploit
Kitploit
•added 2026/09/29 11:27 a.m.•10 views

IPI-exposure-signal

IPI露出シグナル これは、私たちの論文「Your Agentic LLMs Secretly Encode Latent Signals of Indirect Prompt-Injection Exposure」のコヌドリポゞトリです。 ArXiv版ず論文リンク: https://arxiv.org/abs/2608.02657 このリポゞトリは、朜圚的なIPI露出シグナルのプロヌビングパむプラむンAgentDojoトレヌス収集、IPIリスクラベリング、隠れ状態のフィヌチャヌ化、IPI露出プロヌブのトレヌニング/評䟡を実装しおいたす。 匕甚...

6.2AI score
SaveExploits0
Vulnrichment
Vulnrichment
•added 2026/09/29 12:00 a.m.•8 views

CVE-2026-77177

Open GenAI Stack aka ogx-ai 2026-06-11, as used in the Meta AI backend for WhatsApp and other products, allows code execution because prompt injection with Jinja2 template syntax can be used to achieve server-side expression evaluation without sanitization...

6.1AI score0.00592EPSS
SaveExploits0References1
EUVD
EUVD
•added 2026/09/29 12:00 a.m.•8 views

EUVD-2026-89070

Open GenAI Stack aka ogx-ai 2026-06-11, as used in the Meta AI backend for WhatsApp and other products, allows code execution because prompt injection with Jinja2 template syntax can be used to achieve server-side expression evaluation without sanitization...

9.8CVSS6.1AI score0.00592EPSS
SaveExploits0References1
Cvelist
Cvelist
•added 2026/09/29 12:00 a.m.•40 views

CVE-2026-77177

Open GenAI Stack aka ogx-ai 2026-06-11, as used in the Meta AI backend for WhatsApp and other products, allows code execution because prompt injection with Jinja2 template syntax can be used to achieve server-side expression evaluation without sanitization...

0.00592EPSS
SaveExploits0References1
CVE
CVE
•added 2026/09/29 12:00 a.m.•40 views

CVE-2026-77177

The Open GenAI Stack (also known as ogx-ai ) version 2026-06-11 , utilized in the Meta AI backend for WhatsApp and other products, is vulnerable to code execution . The root cause is a lack of sanitization allowing prompt injection using Jinja2 template syntax , which enables server-side expressi...

9.8CVSS6.1AI score0.00592EPSS
SaveExploits0References1
Kitploit
Kitploit
•added 2026/09/28 10:11 p.m.•13 views

skulto

Skulto AI゚ヌゞェントスキルをむンストヌルする前に、プロンプトむンゞェクションをスキャンしたす ╔════════════════════════════════════════════════════════╗ ║ ███████╗██╗ ██╗██╗ ██╗██╗ ████████╗ ██████╗ ║ ║ ██╔════╝██║ ██╔╝██║ ██║██║ ╚══██╔══╝██╔═══██╗ ║ ║ ███████╗█████╔╝ ██║ ██║██║ ██║ ██║ ██║ ║ ║ ╚════██║██╔═██╗ ██║ ██║██║ ██║ ██║ ██║ ║ ║...

6.3AI score
SaveExploits0References1
Rows per page
Query Builder