Secure Coding Drift in LLM-Assisted Post-Quantum Cryptography Development: A Gamified Fix
随着后量子密码学（PQC）的过渡，实现复杂性显著增加，要求严格遵守恒定时间执行、侧信道抗性和精确参数化。同时，大型语言模型（LLM）已深度嵌入软件开发流程，包括密码工程。尽管LLM提高了生产力，但有证据表明它们经常生成不安全或次优的代码，特
2026-06-21 归档
今日共收录 748 条安全情报，包含 44 个 CVE，388 篇安全通告，以及 64 篇研究论文。
← 返回归档列表CodeSentinel: A Three-Layer Defense Against Indirect Prompt Injection in Code Contexts
该论文提出 CodeSentinel，一种针对代码大语言模型（Code LLM）中间接提示注入攻击的三层推理时防御系统。研究背景：代码大语言模型在编程辅助中常从外部仓库、文档、问题线程和编码智能体环境检索代码上下文，攻击者可利用此过程在注释
PhantomSkill: Malicious Code Injection in Agent Skill Ecosystems
本文提出了名为 PhantomSkill 的攻击框架，针对基于大型语言模型（LLM）的编码代理（coding agents）所使用的技能生态系统。此类代理通过安装第三方技能包获取特定领域能力，但这也引入了新的供应链攻击面。PhantomSk
OpenAnt: LLM-Powered Vulnerability Discovery Through Code Decomposition, Adversarial Verification, and Dynamic Testing
大规模代码库的自动化漏洞发现仍然面临挑战：传统静态分析误报率高，而动态方法（如模糊测试）需要大量基础设施且通常针对狭窄的漏洞类别。近年来，大型语言模型（LLM）的进展使得对程序行为进行语义推理成为可能，但将LLM应用于仓库级安全分析会遇到上
Artificial Intelligence as Game Changer in Cybersecurity: What We Learned in 2025-2026, and how this is relevant for Africa
本文基于2025-2026年发生的两起标志性事件，论证了前沿大型语言模型已成为网络行动的决定性工具，且该工具被少数行为体（美国科技公司、盟国政府及欧洲标准机构）所垄断，非洲被完全排除在外。第一起事件中，一个大型语言模型几乎独立执行了一场国家
SurrealDB: Denial of Service via deep operator chains
An authenticated user could crash a SurrealDB server with a single query containing a long chain of operators. Such a q
Lokka: Azure Resource Manager URL path validation issue
Lokka versions prior to 2.1.2 constructed Azure Resource Manager request URLs using direct string concatenation with use
LangSmith SDK TracingMiddleware: Arbitrary server-side file read
# Summary An attacker who can send an HTTP request to a server running the LangSmith SDK's `TracingMiddleware` can caus
githubtoplanguages: Command Injection via Issue Title in Discord Notification Workflow
### Summary A GitHub Actions workflow is vulnerable to command injection through the issue title. The workflow is trig
ChatterBot: Symlink-Following Arbitrary Write via UbuntuCorpusTrainer
## Summary ChatterBot's `UbuntuCorpusTrainer.extract()` uses a predictable, home-rooted output directory (`~/ubuntu_dat
appium-mcp: Unescaped Locator Data XSS in MCP-UI Resource (createLocatorGeneratorUI)
## Unescaped Locator Data XSS in MCP-UI Resource (createLocatorGeneratorUI) ### Summary `appium-mcp`'s `createLocatorG
EverOS: Path traversal in EverOS /api/v1/memory/add via unvalidated sender_id
EverOS versions 1.0.0 and earlier are vulnerable to path traversal in the POST /api/v1/memory/add ingestion endpoint. Th
MessagePack for Python: Out-of-bounds read / crash on Unpacker reuse after a caught error
### Impact If the Unpacker is used repeatedly after an error occurs, the process may crash with a SEGV. If the Unpacke
Gogs: XSS in .ipynb files renderer due to outdated notebookjs
### Summary Gogs renders Jupyter notebook files (`.ipynb`) using [jsvine/notebookjs](https://github.com/jsvine/notebook
parse-server: LiveQuery discloses object data to a subscriber across an ACL read-access change
### Impact A Parse Server LiveQuery subscriber can receive object field values they are not authorized to read when a s
SearXNG MCP Server: DNS-resolved Private Hostname SSRF in `web_url_read`
## DNS-resolved Private Hostname SSRF in `web_url_read` ### Summary The `web_url_read` MCP tool in `mcp-searxng` is vu
SearXNG MCP Server: Unbounded Response Body Read Bypasses URL Size Limit in `web_url_read`
## Unbounded Response Body Read Bypasses URL Size Limit in `web_url_read` ### Summary The `web_url_read` MCP tool in m
Network-AI: EnvironmentManager.restore() backup ID path traversal copies arbitrary directories into environment data
### Summary `EnvironmentManager.restore(env, backupId)` computes the backup path with `join(envDir, '.backups', backupId
Network-AI: EnvironmentManager.backup() follows symlinked directories and copies files outside the environment root into backups
### Summary `EnvironmentManager.backup()` recursively collects files using `_collectBackupFiles()`. `_collectBackupFiles
Network-AI: Poisoned environment backup manifest allows arbitrary recursive deletion during backup pruning
### Summary `EnvironmentManager.listBackups()` reads each backup's `_manifest.json` and trusts the manifest's `path` fie