[FEATURE] Add memory security scanning to prevent memory poisoning in shared crew memory

[vgudur-dev]

Feature Area

Core functionality

Is your feature request related to a an existing bug? Please link it here.

• Manual input validation before passing to crew (doesn't catch encoded attacks)
• Prompt-level instructions to agents to "ignore suspicious content" (easily bypassed)
• External guardrails like NeMo Guardrails (heavy infrastructure, not memory-specific)
• Using OWASP Agent Memory Guard as standalone middleware wrapping CrewAI (works but not native)

Describe the solution you'd like

References:

• OWASP Agent Memory Guard
• OWASP Top 10 for LLM Applications - documents memory poisoning
• PyPI: pip install agent-memory-guard

This is especially critical for CrewAI because:

1. Multiple agents share the same memory pool
2. One compromised agent can poison memory for all others
3. Long-running crews accumulate more attack surface over time

Describe alternatives you've considered

No response

Additional context

No response

Willingness to Contribute

Yes, I'd be happy to submit a pull request