Implements defense-in-depth controls at an AI agent's tool-invocation boundary using tool allowlisting, least-privilege identity binding, NeMo Guardrails policy enforcement, human-in-the-loop approval, and audit logging.
Install
npx skills add mukul975/Anthropic-Cybersecurity-Skills --skill securing-agentic-ai-tool-invocationnpx skills add mukul975/Anthropic-Cybersecurity-Skills --skill securing-agentic-ai-tool-invocation -a claude-code -gnpx skills add mukul975/Anthropic-Cybersecurity-Skills --skill securing-agentic-ai-tool-invocation -a codex -ggemini skills install https://github.com/mukul975/Anthropic-Cybersecurity-Skills.git --path skills/securing-agentic-ai-tool-invocationgh skill install mukul975/Anthropic-Cybersecurity-Skills securing-agentic-ai-tool-invocationgit 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/securing-agentic-ai-tool-invocation ~/.claude/skills/securing-agentic-ai-tool-invocation # Claude Code
cp -R /tmp/mukul975__Anthropic-Cybersecurity-Skills/skills/securing-agentic-ai-tool-invocation ~/.agents/skills/securing-agentic-ai-tool-invocation # Codex, Cursor, Gemini CLI, Copilot, OpenCode