开源 LLM 工程平台：追踪、评测、提示词管理与指标。
SkillSpector (NVIDIA)
Security scanner for AI agent skills. Detect vulnerabilities, malicious patterns, security risks, prompt injection, data exfiltration, and supply-chain risks in Claude Code, Codex, and MCP skills before you install them.
- 仓库
- NVIDIA/SkillSpector
- Stars
- ★ 19,905
- Forks
- 1,744
- 语言
- Python
- 许可证
- Apache-2.0
- 最近更新
- 2026-10-10
- 访问次数
- 0
MCP
同类项目
测试提示词、模型与 RAG 的评测与红队工具，CLI 与 CI 友好。
大模型少样本评测框架，Open LLM Leaderboard 背后的工具。
Open-source AI penetration testing tool to find and fix your app’s vulnerabilities.
用众多可复用的提示词模式增强人类能力的开源框架。
An AI prompt optimizer for writing better prompts and getting better AI results.