docs: Add Inkog security scanning for CrewAI agents

[cloakmaster]

Summary

• Adds a new guide page at docs/en/guides/advanced/security-scanning.mdx documenting Inkog, an open-source security scanner purpose-built for AI agent code
• Inkog has a native CrewAI adapter that understands agent definitions, task delegation, tool usage, and crew orchestration to detect behavioral vulnerabilities that generic linters miss
• Covers CLI usage (npx -y @inkog-io/cli scan .), MCP server setup for Claude Desktop / Cursor, GitHub Actions CI/CD integration, and compliance mapping (EU AI Act, NIST AI RMF, OWASP LLM Top 10)
• Detections include: prompt injection, infinite loops, token bombing, missing guardrails, delegation vulnerabilities in multi-agent systems, SQL injection via LLM, and unsafe code execution

Changes

• docs/en/guides/advanced/security-scanning.mdx (new) -- Full guide page with CLI, MCP, CI/CD, and compliance sections
• docs/docs.json -- Added navigation entry under Guides > Advanced

Test plan

• Verify page renders correctly with mintlify dev
• Verify navigation link appears under Guides > Advanced
• Verify all external links resolve (GitHub, npm)

🤖 Generated with Claude Code

---

Note

Low Risk
Low risk documentation-only change that adds a new page and updates navigation; no runtime code paths are modified.

Overview
Adds a new advanced guide, security-scanning.mdx, documenting how to use Inkog to scan CrewAI projects (CLI usage, MCP/IDE setup, GitHub Actions integration, and compliance mappings).

Updates docs.json navigation to include the new Security Scanning page under Guides → Advanced.

^{Written by Cursor Bugbot for commit 8fde7c0. This will update automatically on new commits. Configure here.}