boaviztapi-mcp
Related Servers
Alternatives to boaviztapi-mcp
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityBmaintenanceEnables agents to estimate the water and energy consumed by chat prompts, a year of usage, or an entire agent session from its token counts, returning everyday comparisons along with the accounting boundary, assumptions and methodology version behind each figure.1MIT
- AlicenseNot gradedqualityBmaintenanceEnables querying measured GPU energy per million output tokens by model and GPU, estimating energy and CO2e for token counts and grid regions, comparing GPUs, exploring concurrency curves, and retrieving hardware and source metadata.329 npmMIT
- FlicenseAqualityDmaintenanceEstimates the environmental footprint of your AI use — energy (kWh), miles driven, water used for cooling, and CO₂ — plus a prompt-efficiency score, working with any AI client by measuring token usage.9-
- AlicenseNot gradedqualityBmaintenanceBridges to the hosted GreenCalculus corpus so clients can look up sourced greenhouse-gas emission factors with full audit envelopes, search or resolve factor keys from plain-language descriptions, and run deterministic calculations for activity data, embodied carbon, PCAF financed emissions, Scope 2 electricity, GLEC freight, spend-based EEIO screening, and business travel.MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI clients to look up Taiwan government emission factors and run traceable, reproducible emission calculations, including electricity, fuel, refrigerant, water, and product carbon footprint data. Also builds scope-categorized inventory summaries and returns factor versions, sources, and usage limits with every result.15 npmMIT

leadgreen-mcpofficial
AlicenseNot gradedqualityBmaintenanceEnables MCP clients to query open data and a JSON API for the energy and CO2e cost of LLM tokens by model, GPU, and country.Apache 2.0
TDQS
Scored across 8 tools
Each tool targets a clearly distinct step or domain: reference lookups (list_impact_criteria, search_catalog, describe_catalog_item), single-item estimates split by resource type (cloud_instance, server, component, device), and cross-option comparison. The descriptions explicitly state the boundary between neighbours ('use compare_estimates to rank several', 'search_catalog to find names'), leaving no realistic chance of misselection.
All eight tools follow a strict verb_noun snake_case pattern, with the estimate_* family forming a predictable, parallel series (estimate_cloud_instance, estimate_server, estimate_component, estimate_device). Catalog tools also pair consistently (search_catalog / describe_catalog_item).
Eight tools is well within the sweet spot and each one earns its place: three reference/lookup helpers, four estimation entry points mapping to distinct physical scopes, and one aggregation tool. Nothing appears redundant or padded.
Coverage of the domain is strong: name resolution, item introspection, criterion glossary, per-scope estimates, and multi-option comparison form a coherent end-to-end workflow with no dead ends for typical questions. Minor gaps remain — there is no direct 'list supported kinds/archetypes' or filter/search over prior results, so agents must probe via search_catalog and per-kind arguments.