Probes Retrieval-Augmented Generation pipelines for indirect prompt injection via poisoned retrieved documents and embedding-space manipulation, using NVIDIA garak, Promptfoo red-team plugins, and Microsoft PyRIT against vector stores like FAISS, Chroma, Pinecone, or pgvector.
安装
npx skills add mukul975/Anthropic-Cybersecurity-Skills --skill testing-prompt-injection-in-rag-pipelinesnpx skills add mukul975/Anthropic-Cybersecurity-Skills --skill testing-prompt-injection-in-rag-pipelines -a claude-code -gnpx skills add mukul975/Anthropic-Cybersecurity-Skills --skill testing-prompt-injection-in-rag-pipelines -a codex -ggemini skills install https://github.com/mukul975/Anthropic-Cybersecurity-Skills.git --path skills/testing-prompt-injection-in-rag-pipelinesgh skill install mukul975/Anthropic-Cybersecurity-Skills testing-prompt-injection-in-rag-pipelinesgit clone --depth 1 https://github.com/mukul975/Anthropic-Cybersecurity-Skills /tmp/mukul975__Anthropic-Cybersecurity-Skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/mukul975__Anthropic-Cybersecurity-Skills/skills/testing-prompt-injection-in-rag-pipelines ~/.claude/skills/testing-prompt-injection-in-rag-pipelines # Claude Code
cp -R /tmp/mukul975__Anthropic-Cybersecurity-Skills/skills/testing-prompt-injection-in-rag-pipelines ~/.agents/skills/testing-prompt-injection-in-rag-pipelines # Codex, Cursor, Gemini CLI, Copilot, OpenCode