MCP App Store
by KeenEthics
EcoCheck icon

EcoCheck MCP Server for Claude and ChatGPT

By EcoCheck by BeevR Technology

Overview

EcoCheck is a greenhouse-gas (GHG) inventory and EU CBAM platform used by manufacturers, exporters and their consultants in Vietnam. With the EcoCheck connector you can do your carbon work by just asking, in English or Vietnamese: - Record monthly activity data (electricity, fuels, refrigerants, purchased goods) for each facility; EcoCheck picks the right emission factor and unit and calculates tCO2e. - Get emissions by scope, ISO 14064-1 category, gas, source and month, find the largest sources, and track reduction targets. - Check whether a product's CN code is covered by the EU Carbon Border Adjustment Mechanism, look up EU default values, and compute actual specific embedded emissions (SEE) for steel, aluminium, cement and fertiliser installations. - Calculate product carbon footprints (ISO 14067, cradle-to-gate or cradle-to-grave) from ready templates or your own processes and inputs, see the main contributors and uncertainty, and export the ISO 14067 report, a draft EPD or a PACT footprint for customers. - Export Word inventory reports (ISO 14064-1 or Vietnam Decree 06/2022 templates). Everything is saved in the same EcoCheck inventory your team uses on the web. Sign-in uses OAuth: Claude never sees your password. The workspace owner sets each member's AI level (view only, view and edit, full access), can turn AI off, and reviews an audit log of every action. Deletions always require explicit confirmation and the Full access level. Documentation: https://ecocheck.ai/ai/docs

How to connect

  1. Claude: Open the EcoCheck listing in the Claude connectors directory and choose Connect. Claude listing EcoCheck exposes 53 tools here.
  2. ChatGPT: Open the EcoCheck listing in the ChatGPT app directory and choose Connect. ChatGPT listing EcoCheck exposes 53 tools here.

Tools

add_accounts_from_template

ChatGPT
Copies system template accounts (with their calculation settings) into a year. Include the parent groups. Ids from lookup_reference template_accounts.

calculate_lca

ChatGPT
Calculates the current model of a product and returns the footprint without saving anything (also on a locked year): kg CO2e per declared unit, stages, processes, top contributions, gases, data quality, other indicators. mode=uncertainty: Monte Carlo (95 % interval) · mode=scenarios: the baseline and every scenario of the product (save_lca_model entity=scenario) · mode=sensitivity: one-at-a-time parameter sensitivity (each parameter ± delta_pct, default 10 %). To keep a result, use save_lca_product action=save_result or action=finalise.

cancel_invitation

ChatGPT
scope=workspace (owner): cancels a workspace invitation (invitation_id from get_workspace_admin). scope=organization / inventory: cancels the invitation sent to email. Ask the user first.

change_member_role

ChatGPT
Sets the role (READ / WRITE) of an organization or inventory member (user_id from get_organizations / get_inventories with include_members). Needs the Full access AI level.

clone_inventory_year

ChatGPT
Copies a year (accounts, calculation settings, assessments, CBAM structure; no activity data) into a new year of the same inventory or of to_inventory_id.

create_inventory_year

ChatGPT
Creates an empty reporting year (status REPORTING, open for data) in an inventory. Use clone_inventory_year to copy the structure of an existing year instead.

delete_lca_records

ChatGPT
Deletes a product footprint (with its model and results), a process (with its inputs), an input, a co-product, a parameter or a scenario (the product must be DRAFT for these), a parameter of the year (global_parameter_id) or an organization template (template_id). Ask the user first.

delete_records

ChatGPT
Deletes: organization, facility (refused while it has journals or CBAM data), inventory, year (with its accounts, journals, CBAM data), emission_account (a group with its children and journals), journals (ids), custom_factor (a value, cfv_…), custom_factor_family (a family, cf_…; delete its values first), target, reduction_action, target_action_link (target_id + id), cbam_process, cbam_good, cbam_precursor, cbam_source_stream. Ask the user first.

ecocheck_guide

ChatGPT
Detailed guidance before acting: journal input per calculation method, CBAM workflow, permissions, data model, Scope 3 categories. Call it when unsure how to fill a write tool.

export_lca_document

ChatGPT
Documents of a saved result (result_id): report = ISO 14067 carbon footprint study report (.docx) · epd = draft Environmental Product Declaration (ISO 14025 / EN 15804+A2, .docx, the basis for third-party verification) · ilcd = ILCD+EPD data set (.zip, for EPD programmes) · pact = PACT v3 ProductFootprint (JSON, FINAL results only). The model of a product (no result_id): workbook = Excel bill of materials (.xlsx) · jsonld = openLCA JSON-LD (.zip) · simapro = SimaPro CSV · ecospold2 = EcoSpold2 (.zip). workbook_template = blank Excel import template (no product_id). lang vi (default) or en for report / epd. Importing files back (Excel, JSON-LD, SimaPro, EcoSpold) is done on ecocheck.ai.

export_report

ChatGPT
kind=inventory: GHG inventory report / certificate (.docx, needs facility_id; counts toward the plan export quota); report_template_code ISO_14064_1 = organisation-wide ISO 14064-1:2018 report (clause 9.3 content, no facility_id). kind=reduction: emission-reduction report (.docx, Decree 06 template RE_ND_06; needs facility_id and target_id). A refusal listing missing_fields names the profile fields to complete first.

fetch

ChatGPT
Full content of a search result: an organization, a facility, or an inventory with its years and latest emissions summary.

finalise_cbam_report

ChatGPT
Freezes the CBAM data of the installation-year (status FINAL): further edits are refused until reopen_cbam_report. Refused while the result has ERROR warnings. Confirm with the user first.

get_assessments

ChatGPT
Per-account significance or data-accuracy scores of a year. Significance template ISO14064_2018 scores ISO 14064-1 5.2.3 criteria (magnitude, influence, data_access, risk_opportunity, sector_guidance, outsourcing, employee_engagement), 0-3 each.

get_cbam

ChatGPT
view: installations (facilities as CBAM installations + status) · data (profile, processes, CN goods, precursors, source streams) · result (specific embedded emissions per CN code, Reg (EU) 2025/2547, AR5, default-value comparison, warnings) · saved_result · default_value (cn_code, country): exact CN8 code from the customs declaration; when the Act prints no row for it, candidates lists the printed codes with the same prefix — pick the matching one, do not substitute another heading.

get_emission_accounts

ChatGPT
Chart of emission sources of a year (tree by code / parent_code). type=LEAF lists the accounts activity data can be recorded on. With account_id: one account with its calculation method and emission factor.

get_emissions_report

ChatGPT
Totals of an inventory year in t CO2e: by scope, source, facility, Scope 3 / ISO 14064 categories, excluded and biogenic. summary.market_based: Scope 2 location-based vs market-based (contractual instruments such as I-REC / DPPA; grid factor as residual mix). summary.consolidation: totals by the organisation's consolidation approach (operational / financial control, equity share). summary.base_year: base-year recalculation status (cumulative change vs threshold).

get_facility

ChatGPT
A facility (plant, site, building) with address and technical blocks.

get_inventories

ChatGPT
Without inventory_id: list inventories (optionally of one organization). With inventory_id: details (GWP set, reporting purpose), reporting years with status, linked facilities and (include_members) members.

get_journals

ChatGPT
Recorded activity data of a year with computed emissions (kg). With journal_id: one entry and its evidence files.

get_lca

ChatGPT
Product carbon footprints (ISO 14067, cradle-to-gate or cradle-to-grave) of an inventory year. view: products (every product with its latest saved result, kg CO2e per declared unit, stages, top contributions) · product (the model: processes, inputs with factor_key / amount, parameters, issues; needs product_id) · results (saved results of a product) · result (one saved result: totals, stages, processes, top contributions, gases, data quality, uncertainty, other indicators; needs result_id) · checks (rule check of a saved result against standard iso14067 | en15804 | pef) · compare (product_ids, first = base, per declared unit, nothing stored). Needs the LCA feature of the plan.

get_my_access

ChatGPT
The signed-in user (id, e-mail, workspaces with role) and their AI-assistant access level in a workspace (NONE / READ / WRITE / FULL, set by the workspace owner). Use it to explain refused actions.

get_organizations

ChatGPT
Without organization_id: list organizations of the workspace. With organization_id: details, facilities, inventories and (include_members) members and pending invitations.

get_targets

ChatGPT
Without target_id: targets (optionally of a year) and the reduction-action catalogue. With target_id: the target and its linked actions.

get_workspace_admin

ChatGPT
Owner only: plan, resources and limits; with include_members, the members (id, e-mail, role, status) and pending workspace invitations.

import_cbam_from_inventory

ChatGPT
Copies Scope 1 CO2 / N2O of the facility journals into CBAM source streams of the installation (CH4 is not a CBAM gas). Re-running keeps the process attributions.

invite_expert

ChatGPT
Sends an invitation e-mail to an external expert (consultant, verifier) for an organization, limited to the given inventories. Needs the Full access AI level. Confirm with the user first: e-mails cannot be recalled.

invite_member

ChatGPT
Sends an invitation e-mail. scope=workspace (owner): invite to the workspace. scope=organization / inventory: invite with role READ or WRITE. Needs the Full access AI level. Confirm the address with the user first: e-mails cannot be recalled.

list_workspaces

ChatGPT
Workspaces (subscriptions) of the signed-in customer: name, subscriptionCode, role. Call first: every other tool needs workspace.

lookup_lca_library

ChatGPT
kind: factors (search emission factors / unit processes of the background databases by words, query required; use the returned key as factor_key of a BACKGROUND input) · dataset (one background dataset by key) · indicators (impact indicators besides GWP) · sources (background databases, licence, attribution) · templates (built-in product templates: metal-parts, garment, plastic-packaging, wood-furniture, agri-food, steel-rolling, aluminium-die-casting, corrugated-box; and the organization's own) · elementary_flows (substances for site EMISSION inputs, query required) · cbam_goods (CBAM goods of the year with their specific embedded emissions, to reuse in a footprint) · facility_inventory (GHG inventory totals of a facility in the year, facility_id).

lookup_reference

ChatGPT
Lookups for the write tools: emission_factors (search the system library: every word of query in any order across names (EN / VI), activity id and description; ranked region (asked, else VN) > official Vietnamese sources > other official sources; filters unit_type / region / year / source), emission_factor (id), custom_emission_factors, account_templates (query: words of the EN or VI name / description, accents optional, e.g. "nam 2026" or "det may" also find the Vietnamese names), template_accounts (id = template; compact rows: code, name, scope / category, method, factor id + values + unit + source code, settings; paged, default 100 per page, meta.hasNextPage; factor detail via emission_factor id), calculation_methods, gwp (code e.g. ar5_100), units, unit_types, sectors, factor_categories, lca_methods, factor_sources (official registry: code, publisher, dataset, edition, licence, attribution, row counts), custom_factor_sources (source codes accepted by save_custom_factor values), factor_regions, factor_years, countries, cities (id = country), certificates, report_templates, assessment_templates, accuracy_templates, reduction_categories, reduction_sub_categories, organization_types, facility_types, inventory_types, enum (code = enum name), scope3_metadata (Scope 3 categories, methods with their input keys, recommended methods, airports, currencies), scope3_factor_sets (code = Scope 3 category 1-15: recommended official factor sets with row counts and sample rows; region = supplier country ranked first, e.g. VN; suspect rows have a lost decimal point).

preview_emission

ChatGPT
kg CO2e of a quantity on a LEAF account, nothing saved. The quantity is in the unit of the account's emission factor (EF.102, e.g. kWh), or in the account's own input unit when it converts (EFC.102: unit_conversion.unit_conversion, e.g. litres of diesel, never pre-converted); for a refrigerant source (Air.101) it is the kg of refrigerant released, valued with the inventory's GWP set.

record_activity

ChatGPT
Books activity data on a LEAF account; EcoCheck computes the emissions. input per method (get_emission_accounts account_id shows calculation_code): EF.102 {value, entry_unit} (e.g. {"value": 1200, "entry_unit": "kWh"}), EFC.102 {value} in the own input unit of the account (unit_conversion.unit_conversion, e.g. litres of diesel — never pre-converted), DI.101 {value} (kg CO2e), Air.101 {amount_val}, S3.SPEND {amount, currency, spend_year}, Scope 3 methods S3.TRAVEL {trips}, S3.TRANSPORT {legs}, S3.COMMUTE {headcount, working_days, modes}, S3.ASSET, S3.PROCESSING, S3.HYBRID, S3.SUPPLIER, S3.FERA, S3.USE, S3.WASTE, S3.PCAF (inputs with examples: ecocheck_guide scope3_categories). Optional data quality fields dq_technology … dq_reliability (1-5), data_type, uncertainty_percent. Dates inside the year; a period over two years is split automatically. dry_run=true only previews. An identical entry (same account, facility, period and input) already recorded is not booked again: the answer is skipped_duplicate with the existing journal; book it anyway only if the user confirms it is a separate entry (allow_duplicate=true).

remove_member

ChatGPT
scope=workspace (owner): removes the member from the workspace and its organizations. scope=organization / inventory: removes the member from it. Ask the user first.

reopen_cbam_report

ChatGPT
Moves a finalised CBAM installation-year back to DRAFT so its data can be edited.

resend_invitation

ChatGPT
Sends the pending organization or inventory invitation e-mail again. Needs the Full access AI level.

save_assessment

ChatGPT
items = significance [{id, assessments: [{criteria_code, value {value}}]}] or accuracy [{id, accuracies: [{criteria_code, value {…}}]}] (ids from get_assessments).

save_cbam_data

ChatGPT
entity=installation: English profile (data = {name_en, country_code, unlocode, latitude, longitude, registry_installation_id, operator_name_en, contact_email…}). entity=process: no id = create (data = {name, category, activity_level_t, route?, electricity_consumed_mwh?, electricity_ef_source?, heat_?, waste_gas_?, params?}), id = update. entity=good (process_id): CN code (data = {cn_code, name_en?, quantity_t?}). entity=precursor (process_id): {source INTERNAL + source_process_id | PURCHASED + cn_code, origin_country?, verified?, see_direct?, see_indirect?, quantity_t} (with id: full replacement). entity=source_stream: {name, method COMBUSTION|PROCESS|MASS_BALANCE|MEASUREMENT, process_id?, gas?, activity_data?, ncv? (TJ / unit), emission_factor?, oxidation_factor?, carbon_content?, biomass_fraction?, measured_t?}. Categories: CALCINED_CLAY, CEMENT_CLINKER, CEMENT, ALUMINOUS_CEMENT, NITRIC_ACID, AMMONIA, UREA, MIXED_FERTILISERS, HYDROGEN, SINTERED_ORE, PIG_IRON, FERRO_MANGANESE, FERRO_CHROMIUM, FERRO_NICKEL, DRI, CRUDE_STEEL, IRON_STEEL_PRODUCTS, UNWROUGHT_ALUMINIUM, ALUMINIUM_PRODUCTS

save_cbam_result

ChatGPT
Stores the current CBAM result of the installation-year as a draft snapshot (get_cbam view=saved_result reads it back).

save_custom_factor

ChatGPT
Workspace admin only (space owner; members get 403): the workspace's own factors. System library factors are maintained by EcoCheck. Deleting a value used by an emission source is refused (409 FACTOR_IN_USE). kind=family, no id: create a factor family (data = {sector_code, category_code, lca_method_code, organizationIds [], name?}); kind=value, no id: add a value (data = {meta_code, source_code (lookup_reference custom_factor_sources, not factor_sources), year_code, region_code, unit_code, supported_calculation_methods [], co2e_val? or co2_val / ch4_val / n2o_val}); with id: update.

save_emission_account

ChatGPT
No account_id: create (data = {name, year_id, type GROUP|LEAF, parent_code e.g. "S1.001", scope3_category? 1-15, calculation_method_instance? for LEAF {calculation_code "EF.102", data {emission_factor: <row of lookup_reference emission_factors> + factor_unit = unit.code + system_factor_value_ref_id = id}}}). With account_id: update fields, or set excluded (true / false + exclude_reason) to move it out of / back into the boundary.

save_facility

ChatGPT
No facility_id: create (data = {organizationId, name 5-255, type OFFICE|PLANT|OTHER, address {street, zip, province, countryId, cityId}}). With facility_id: update.

save_inventory

ChatGPT
No inventory_id: create (data = {organizationId, name 5-255, type CORP|LCA|PROJECT, facilityIds [], interestedCertIds [] REQUIRED, at least one (lookup_reference certificates, e.g. ISO 14064-1 / GHG Protocol), reportingPurpose? COMPLIANCE_ND06|CBAM|VOLUNTARY (sets GWP AR5/AR6), years? [{code:"2025", isBaseYear:true}]}). With inventory_id: update (name, type, gwp, reportingPurpose, facilityIds…).

save_lca_model

ChatGPT
Processes, inputs, co-products, parameters and scenarios (the product must be DRAFT). entity=global_parameter: a parameter shared by every product of the year (no product_id needed; data = {name, value, unit?, description?}; global_parameter_id to update). entity=scenario: a named set of parameter values to compare (data = {name, values {parameter: value}, description?}; scenario_id to update); calculate_lca mode=scenarios compares them. entity=coproduct: a co-product of a process, for allocation (process_id; data = {name, amount or amount_formula, unit, mass_kg_per_unit?, price_per_unit?}; coproduct_id to update; the process allocation_method PHYSICAL | ECONOMIC | MANUAL is set with entity=process). entity=parameter: the quantities of a template model are parameters (get_lca view=product lists them with unit and meaning, all 0 in a new product): data = {name, value} sets one by name (or parameter_id + data); data.values = {name: value, …} sets several at once; a new name creates it (data = {name, value, unit?, description?, formula?}). entity=process: no process_id = create (data = {name, unit, ref_amount?, is_reference?, stage? RAW_MATERIAL|PRODUCTION|DISTRIBUTION|USE|END_OF_LIFE, module? A1…D, mass_kg_per_unit?}), with process_id = update; background_key = copy a background unit process into the model. entity=input: process_id + no input_id = create (data = {kind BACKGROUND (factor_key from lookup_lca_library factors) | FOREGROUND (input_process_id = another process of the model) | EMISSION (gas co2|ch4|ch4_fossil|n2o|co2e or flow_uuid), amount or amount_formula, unit, label?, uncertainty_pct?, dqr_te/dqr_ge/dqr_ti 1-5?}), with input_id = update; input_id + background_key = turn that background input into an editable process of the model (its unit process copied). entity=derived_input: process_id + source facility_inventory (data = {facility_id, scopes [1,2,3], share_percent, output_quantity}) | cbam (data = {facility_id, cn_code, quantity_t}) | cff (Circular Footprint Formula of a material) | supplier_pact (data = {footprint: PACT JSON, quantity}).

save_lca_product

ChatGPT
action=create: new product of the year; template = a built-in template key or template_id = an organization template (lct_…), both from lookup_lca_library templates, to start from a ready model (then set its parameters with save_lca_model entity=parameter), or data = {name, unit (e.g. kg, piece, m2), demand_amount, description?, system_boundary? CRADLE_TO_GATE|CRADLE_TO_GRAVE, method_gwp? ar6|ar5, facility_id?} for an empty model; name is required either way. action=update: product_id + data (same fields; doc_en {product, description, company, facility, address, activity, processes {lcq_…: name}} = English names for English documents; epd_declared {PERM, PENRM, SM, RSF, NRSF, HWD, NHWD, RWD, CRU, MFR, MER, EEE, EET} = declared resource / waste indicators of the EPD). action=clone: product_id + data {name, variant_note?} = an ecodesign variant. action=save_template: product_id + data {name, description?} saves the model as a template of the organization. action=save_result: saves the current calculation as a result (note?). action=finalise: FINAL saves a result and freezes the model (needed for PACT); action=reopen: back to DRAFT.

save_organization

ChatGPT
No organization_id: create (data = {name 5-255, type SME|GOV|CORP|NGOV|EDU|INS|NON_PROFIT_ORG|OTHER, address {street, zip, province, countryId, cityId}, sectorIds [], mainContact {firstName?, lastName?, phone?, email?}, interestedCertIds?}). With organization_id: update the given fields.

save_reduction_action

ChatGPT
No action_id: create (data = {title, subCategoryId (lookup reduction_sub_categories), description?, reduceEmissionPotential?, inventoryEmissionAccountId?, quantity?}). With action_id: update.

save_target

ChatGPT
No target_id: create (data = {title, baseYear, targetYear, annualEmissionReduction %, totalEmissionReduction %, scope ["SCOPE1",…], inventoryYearId, baseYearEmission, nextYearEmission, targetYearEmission}). With target_id: update data, toggle_status (ON TRACK <-> INACTIVE) and / or action_ids (replace the linked reduction actions).

set_inventory_year_status

ChatGPT
REPORTING = open for data (INIT is accepted and stored as REPORTING). LOCKED / FINISHED, or unlocking, need the Full access AI level and the OWNER role; a workspace that requires review refuses LOCKED (REVIEW_REQUIRED: submit for review in the app); a locked year refuses data changes. Confirm with the user before locking or finishing a year.

set_inventory_year_templates

ChatGPT
Sets the significance and / or data-accuracy assessment template of a year (ids from lookup_reference assessment_templates / accuracy_templates).

set_workspace_member_status

ChatGPT
Workspace owner only, Full access AI level: ACTIVE or INACTIVE (an inactive member cannot sign in to the workspace). Confirm with the user before deactivating someone.

update_journal

ChatGPT
Updates a journal and recalculates. data: emission_account_id, facility_id, startBookDate, endBookDate, input, source, data_quality, note, dq_technology, dq_time, dq_geography, dq_completeness, dq_reliability (1-5), data_type (PRIMARY | SECONDARY), uncertainty_percent.

add_accounts_from_template

Claude

calculate_lca

Claude

cancel_invitation

Claude

change_member_role

Claude

clone_inventory_year

Claude

create_inventory_year

Claude

delete_lca_records

Claude

delete_records

Claude

ecocheck_guide

Claude

export_lca_document

Claude

export_report

Claude

fetch

Claude

finalise_cbam_report

Claude

get_assessments

Claude

get_cbam

Claude

get_emission_accounts

Claude

get_emissions_report

Claude

get_facility

Claude

get_inventories

Claude

get_journals

Claude

get_lca

Claude

get_my_access

Claude

get_organizations

Claude

get_targets

Claude

get_workspace_admin

Claude

import_cbam_from_inventory

Claude

invite_expert

Claude

invite_member

Claude

list_workspaces

Claude

lookup_lca_library

Claude

lookup_reference

Claude

preview_emission

Claude

record_activity

Claude

remove_member

Claude

reopen_cbam_report

Claude

resend_invitation

Claude

save_assessment

Claude

save_cbam_data

Claude

save_cbam_result

Claude

save_custom_factor

Claude

save_emission_account

Claude

save_facility

Claude

save_inventory

Claude

save_lca_model

Claude

save_lca_product

Claude

save_organization

Claude

save_reduction_action

Claude

save_target

Claude

set_inventory_year_status

Claude

set_inventory_year_templates

Claude

set_workspace_member_status

Claude

update_journal

Claude

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/ecocheck"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/ecocheck)