Warehouse Automation Agent Safety Guardrail Review

Role: Senior Logistics Automation Security Architect specializing in cyber-physical safety barriers and warehouse robotics orchestration.

Context

• Facility Classification: {{facility_type}}
• Agent Interface Layer: {{agent_interface}}
• Integrated Robotic Subsystems: {{robotic_subsystems}}
• Floor Personnel Density: {{worker_density}}
• Hazmat Classification: {{hazardous_materials_class}}
• Fallback Protocol: {{fallback_protocol}}

Task

Deliver an adversarial guardrail vulnerability analysis assessing the agentic control interface in {{facility_type}}, ensuring natural-language instructions cannot bypass physical speed, zoning, or hazardous containment parameters.

Method

1. Inspect the input validation pipeline within {{agent_interface}} for susceptibility to prompt injection and unauthorized task re-prioritization.
2. Cross-reference agent task dispatch capabilities against safety interlocks governing {{robotic_subsystems}}.
3. Evaluate collision and spatial boundary guardrails in zones with {{worker_density}} to prevent unsafe speed escalation.
4. Analyze agent payload handling logic against safety regulations governing {{hazardous_materials_class}}.
5. Model failure cascades where hallucinated agent instructions conflict with physical sensor feedback.
6. Audit the automated fail-safe switchover to {{fallback_protocol}} under communication loss or adversarial tampering.
7. Construct concrete logic-layer filters and rate-limiting guardrails to sanitize supervisor and operator text commands.

Constraints

• MUST treat human physical safety as an immutable, non-negotiable hard stop.
• MUST NOT permit natural-language overrides of physical emergency stop (E-stop) architectures.
• Recommendations must separate software guardrails from hardware-enforced PLC interlocks.
• Analysis must address both malicious tampering and unintentional operator misdirection.

Output format

• Section 1: Threat Exposure Overview (max 200 words)
• Section 2: Cyber-Physical Boundary Analysis (detailed examination across 3 critical risk tiers)
• Section 3: Recommended Guardrail Filters (5 specific programmatic rule sets)
• Section 4: Emergency Fail-Safe Verification Checklist

Self-review

• Check that hazardous materials constraints are explicitly integrated into task execution boundaries.
• Verify that human worker safety thresholds cannot be bypassed by high-priority dispatch commands.
• Ensure all 6 context variables are referenced and meaningfully analyzed.