Loading ToolBench
Mission-critical data center design & engineering MCP server. Provides cooling load calculations, power redundancy analysis, tier classification assessment, and commissioning workflow tools for AI agents.
Strong domain-specific tool suite with comprehensive parameter schemas and detailed descriptions. All 8 tools have clear names following verb_noun patterns (calculate, analyze, assess, generate, lookup). Descriptions are technical and thorough (150-400 chars), well above minimum. All input schemas use Zod with proper type definitions and enum constraints. However, OUTPUT schemas are not explicitly documented, responses are inferred from tool descriptions rather than formally specified. Error handling exists but lacks recovery guidance (no 'if this fails, try X' patterns). Security is solid (READ_ONLY annotation on all tools, no credentials in params). Tool composition is clean with single responsibilities, though some tools could benefit from clearer chaining guidance.
Analyze power redundancy configuration for a mission-critical data center. Calculates UPS module count and sizing, generator count and total capacity, PDU requirements and switchgear feeds, electrical efficiency chain losses, UPS loading percentage, and concurrent maintainability / fault tolerance assessment. Supports N, N+1, 2N, and 2N+1 redundancy configurations.
Analyze rack power density and cooling requirements for a data center deployment. Classify rack density and recommend cooling strategy, containment, and airflow.
Assess a data center's Uptime Institute Tier classification based on its infrastructure configuration. Evaluates power redundancy, cooling redundancy, distribution paths, concurrent maintainability, and fault tolerance against Tier I-IV requirements. Identifies gaps between current infrastructure and target tier.
Calculate the total cooling load for a data center facility based on IT load, PUE, and environmental factors. This tool computes cooling capacity requirements in kW, tons, and BTU/hr, estimates required airflow in CFM, rates the facility's PUE efficiency, and provides actionable recommendations for cooling system design. Accounts for: IT heat rejection, electrical overhead losses, lighting loads, humidification, and altitude derating above 5,000 ft.
Output schemas not formally documented. Tool descriptions mention what is returned (e.g., 'computes cooling capacity requirements in kW, tons, and BTU/hr'), but the exact JSON structure users should expect is not specified in a machine-readable schema.
Error handling lacks recovery guidance. Tools use toolResult() wrapper that catches errors and returns structured error text, but error messages do not suggest next actions (e.g., 'if altitude is out of range, check equipment derating tables').
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | C | 68 | 2025-06-18+ | v2 |
| 2026-03-09 | C | 69 | - | v1 |
Generate a comprehensive data center commissioning plan following ASHRAE guidelines (Levels 1-5). Creates a structured commissioning plan with test procedures, durations, milestones, and prerequisites scaled to facility size and tier complexity. Commissioning Levels: - L1: Factory Witness Testing - L2: Component Verification & Startup - L3: System Functional Performance Testing - L4: Integrated Systems Testing (load bank, concurrent maintenance, fault injection) - L5: Operational Sustainability (seasonal verification)
Model GPU cluster cooling: load, strategy, coolant flow, CDUs, and annual energy cost delta. Specialized tool for high-performance computing environments. Calculate power requirements and cooling strategies for GPU-intensive workloads with precision.
Look up tier requirements, PUE benchmarks, rack density classes, and commissioning phases.
Size UPS modules and battery strings for target runtime and redundancy level. Size UPS systems and battery banks based on load profiles, desired runtime, and efficiency curves. Include aging factors and environmental conditions.
dc_reference_lookup tool name is generic. 'lookup' is a weak verb; better names would be 'get_tier_requirements', 'search_reference', or 'describe_reference_data' to clarify intent without reading the description.
Some parameter descriptions lack explicit format/range guidance. E.g., 'pue' (Power Usage Effectiveness) is described as 'ratio of total facility power to IT power (typically 1.2-1.8)' but does not explicitly state valid range constraints or validation rules in the Zod schema visible in source.
Tool composition could be clearer. dc_analyze_rack_density and dc_calculate_cooling_load both deal with cooling but operate at different levels (per-rack vs facility). Guidance on when to call which tool (e.g., 'call dc_calculate_cooling_load for facility-level planning, dc_analyze_rack_density for rack-level deployment') is absent.