mcp · Global

ghg-calculator

GHG Protocol emissions calculator with 967 embedded factors, CLI, MCP server, and Claude Code skill

Claim this listing

Represent ghg-calculator? Send a claim for review so we can discuss listing updates. Claims do not automatically grant ownership or verification.

Data offered docs-verified: Scope 1 emissions (stationary/mobile combustion, fugitive, process) · Scope 2 purchased electricity, location-based and market-based · Scope 3 across all 15 value-chain categories · 967 embedded emission factors (EPA Hub, eGRID, DEFRA, USEEIO, Ember, EXIOBASE) · GWP lookup (AR5 default, AR6 supported) · Unit conversion (energy, mass, volume, distance) · Interactive HTML carbon reports with Plotly charts
Coverage llm-inferred: United States (US eGRID subregions), 120 countries for electricity (Ember), UK transport/materials (DEFRA), and EU, CN, JP, IN, BR, RU multi-regional (EXIOBASE)
Resolution llm-inferred: Electricity at US eGRID subregion level (27) or country level for international Ember factors; other factors at category or NAICS-sector level
Formats docs-verified: JSON (activity records) · HTML (interactive report) · CLI · Python API · MCP server
Access docs-verified: Free and open source, MIT licence (docs); no API key or paid service required; local install via git clone and uv sync, requires Python 3.11+. Entry licence field is null, so it should be updated to MIT.

Use cases llm-inferred: Scope 1 to 3 carbon footprint for a UK consultancy project using DEFRA and Ember UK grid factors · Scope 1 site-works emissions from plant and fleet diesel records · Embodied carbon screening of construction materials using DEFRA material factors · Client GHG inventory for ESG or planning-stage carbon reporting
Audience llm-inferred: environmental consultants · civils and construction sustainability teams · ESG consultants · climate tech developers

Constraints llm-inferred: Carbon accounting only; no water, flood, hydrology or ground data · Spend-based Scope 3 uses USEEIO, which is US NAICS-based, so UK spend coverage is weak · Factor vintages and database version dates are not stated in the docs · Entry tools list includes Python dependencies (pydantic, pint, pandas, plotly, fastmcp, jinja2) that are not MCP tools; the documented MCP tools are the 8 named in the docs · Entry licence is null but docs state MIT · Requires a local Python 3.11+ and uv install; no hosted endpoint is documented · Accuracy depends on input data quality (docs)

Licence: Unknown — check provider terms
Authentication: No API key required (repository README, auto-extracted; check upstream requirements)
GitHub stars: 13
Last commit: 17/02/2026
Last successful page check: 14/10/2026
Page health: healthy
Package: Not recorded
Verification: Not independently verified

Auto-extracted, unverified fields: tools, install, auth. Check upstream documentation.

Tools/install/auth auto-extracted from README; quality score heuristic (stars, recency, tool count, licence).

Metadata observed 15/10/2026 · research sweep: github-search