作者 | 仇智慧策划 | 李冬梅随着 Anthropic 开源 skills 仓库，"Code Interpreter"（代码解释器）模式成为 Agent 开发的热门方向。许多开发者试图采取激进路线：赋予 LLM 联网和 Python ...

Three serious prompt injection vulnerabilities in Anthropic’s Git MCP server briefly enabled remote code execution and file ...

SciToolAgent is a powerful agent framework designed to integrate diverse scientific tools with large language models (LLMs) to address the limitations of existing systems in scientific research. By ...

Anthropic, the company behind the popular AI model Claude has fixed multiple bugs in its Git MCP server which, researchers ...

High-severity flaws in the Chainlit AI framework could allow attackers to steal files, leak API keys & perform SSRF attacks; ...

Three vulnerabilities in Anthropic’s MCP Git server allow prompt injection attacks that can read or delete files and, in some ...

Vulnerabilities in Chainlit could be exploited without user interaction to exfiltrate environment variables, credentials, ...

Two "easy-to-exploit" vulnerabilities in the popular open-source AI framework Chainlit put major enterprises' cloud ...

Security researchers uncovered two vulnerabilities in the popular Python-based AI app building tool that could allow ...

The Python-based information stealer SolyxImmortal uses legitimate APIs and libraries for stealthy data gathering and ...