South Africa · Cookstoves (VMR0006) · Read by the checks: Project description + 3 monitoring reports + verification report + validation report, 450 pages · Verra registry page
Signal, not verdict. These findings summarize what the project documents say and where. They are not a rating, a certification or an ICVCM assessment, and they can be wrong.
Looked up by methodology code in table version 2026-10-15. No model is involved, and this status is kept apart from the AI-assisted findings below.
0xc459…b488(opens Sepolia Etherscan in a new tab)version 6; earlier versions stay readabledeepseek/deepseek-v4.1-flashtypesafe/jev-1.13Third-party models, not developed by this team.The document’s own figures against registry records (CarbonPlan OffsetsDB, 2026-06-01). A flag is a lead for a reviewer, not a finding of wrongdoing.
No table of yearly net emission reductions could be read from the documents.
Issued credits per vintage over monitored net reductions (2023-2024): 2023: 2.05, 2024: 13.25. 0 of 2 vintages are at or below the monitored figure (a ratio below 1 is expected where a buffer share is withheld). Above it: 2023, 2024, a vintage the monitoring reports read cover only in part, so a later period may hold the rest.
Across 5 vintages, retired volume never exceeds issued volume.
Document and registry both give VMR0006.
No project in the same country with a closely matching name.
Every number, page and table read is in checks/VCS2505.json.
AI-assisted checks: 11 of 15 found evidence in the excerpt read for their principle, each citing the paragraph it came from. 1 check the models disagreed on was decided by a team member, whose answer counts; the models’ answers are shown next to it.
Tested on 177 labelled questions from 11 projects (short 2,500-character excerpts), the checks were right 94% of the time (95% interval 89%–96%). Reports read longer excerpts, which find more: of 61 findings the short excerpts missed, 46 were confirmed (75%, judged by an AI reviewer, who agreed with a person on 18 of 19 spot checks). How this was measured.
Robustness: deleting the paragraphs that held the evidence turned 87% of “found” answers to “not found”, and unrelated text got “not found” on 177 of 177 questions. A paragraph planted in the document telling the model to answer “yes” changed 0 answers; one telling it to answer “no” changed 71%, which is why such paragraphs are screened out before a model reads the excerpt (the screen is a pattern match and can miss new wordings). Robustness tests.
Excerpt read: 29,685 characters chosen by keyword, not the whole document
Evidence found for 1 of 4 checks (A3); not found in the excerpt: A1, A2, A4; 1 decided by a person after the models disagreed.
Decided by a team member: not found (checklist model: not found; second model 62% likely found)
“1. There is no government mandated programme or policy in South Africa ensuring the distribution of new project activity instances or fuel efficient cookstoves.”
Emission-avoidance activity without stored carbon at risk of reversal; permanence is not applicable (rule based on project category, not AI).
Not scored for this activity type (no reversal risk).
Excerpt read: 44,854 characters chosen by keyword, not the whole document
Evidence found for 4 of 4 checks (Q1, Q2, Q3, Q4).
“Bold tonnes/year 5.49 Quantity of biomass consumed in baseline”
“Efwf,CO2 tCO2e/TJ 112 CO2 emission factor of biomass that is substituted or reduced”
“Efwf,non-CO2 tCO2e/TJ 26.23 Non-CO2 emission factor of biomass that is substituted or reduced”
“DF fraction 0.95 Default leakage adjustment factor”
“Comments Parameter Bold once determined shall remain fixed for the entire crediting period. Where charcoal is used by the baseline devices, a default wood to charcoal conversion factor of 6 kg of firewood per kg of charcoal may be used in line with paragraph 35 of AMS II.G, version 11.”
“Description Fraction of woody biomass that can be established as non- renewable biomass”
“Source of data Calculated based on the CDM TOOL 30 “Calculation of the fraction of non-renewable biomass” v3.0 for PI1-K2C. See associated report: Biomass Baseline for Cookstove Project in the Kruger to Canyon and Waterberg Biospheres;”
“Description CO2 emission factor of biomass that is substituted or reduced.”
“Source of data VMR0006 methodology default: IPCC default for wood fuel”
“Description Non-CO2 emission factor of biomass that is substituted or reduced.”
“Source of data VMR0006 methodology default: IPCC default for wood fuel”
“The project adopts a gross-to-net adjustment factor of 95% on emissions reductions to account for leakage as described under the methodology VMR0006, Section 8.3.”
“Leakage shall be considered as default 0.95 in accordance with Section 8.3 of the applied methodology VRM0006 v1.”
“0.95 Discount factor to account for leakage.”
“A net gross leakage factor of 0.95 is applied to ERy,i,j.”
“DF fraction 0.95 Default leakage adjustment factor”
“from PI1-K2C. Based on the initial number of ICS planned for PI1-K2C and taking the expected 5-year operational lifetime of the ICS into account, the GP’s ex-ante ERs were modelled to represent a conservative approximation. Although the expected lifetime of the technology is 5 years, it is assumed that some of the devices will still be operational after the 5-year period. For ex-ante estimations, a decreasing operation rate of stoves (“drop off rate”) of 2.5% per year was incorporated into the ER calculation for the 10-year crediting period.”
“0.94 Adjustment factor to account for uncertainty related to project cookstove efficiency test.”
“Data from the monitoring survey will be collected at least biennially and applied to the emissions reductions calculations during that monitoring period. Where the required level of precision is not met either additional surveys may be completed, or the lower bound value applied.”
“Where the required level of precision is not met either additional monitoring activities may be completed, or the lower bound value applied.”
“Where the required level of precision is not met either additional monitoring activities may be completed, or the lower bound value applied.”
“Data from the monitoring survey will be collected at least once every two years and applied to the emissions reductions calculations during that monitoring period. Where the required level of precision is not met either additional surveys may be completed, or the lower bound value applied.”
“If the required level of precision is not achieved, then further surveys may be completed or the lower bound value adopted in the calculations.”
“The total number of devices in operation compared to the total number of distributed (according to the monitoring database) will be surveyed.”
“Data from the monitoring survey will be collected at least biennially and applied to the emissions reductions calculations during that monitoring period. Where the required level of precision is not met either additional surveys may be completed, or the lower bound value applied.”
“Monitoring equipment When measured: as specified in WBT protocol”
“Monitoring equipment As specified in KPT protocol”
“Data from the monitoring survey will be collected at least once every two years and applied to the emissions reductions calculations during that monitoring period. Where the required level of precision is not met either additional surveys may be completed, or the lower bound value applied.”
“Field-based monitoring activities to determine monitored parameter values will be either a field survey, KPT or WBT.”
“based on the information included in the final monitoring report, there are several procedures for data collecting depending on the methodology applicable for each step of the project. The organizational structure/26/, along with roles and responsibilities throughout project implementation, has been provided in the MR and verified. PP conducted sampling surveys to gather information needed for the monitoring of Ny,i,j and μy parameters. Monitoring surveys/21/22/23/ were conducted across both PIs between June and July 2024. The minimum sample size of 100 households as per VMR0006 Section 8.4 Option (b) were selected to survey for the proportion parameter ny,i,j.and µy. Sampling captured information on monitoring variables with required confidence/precision (90/10 for individual PIs on an annual basis, or 95/10 confidence/precision basis for groups of PIs or sampling on a biennial basis). PP has applied 90/10 confidence/precision for sampling for each individual project instance. A simple random sampling was used. The applied confidence interval and precision level is 90/10. The sample size for the parameter μy is determined based on the section 8.4 Option (b) of the. VMR0006 methodology, i.e. Project target population > 1000: Minimum sample size 100. PP has applied a sample size of 101 for K2C, 106 for WB and 105 for VH, both parameters Ny,i,j and μy. The sampling approach has been demonstrated in the monitoring survey/21/22/23/ for each PIs. Relative precision of data is calculated at 8.2% for K2C, 0% for WB and 7% for VH determined for the parameter ny,i,j (which is used to calculate parameter Ny,i,j and is less than the precision level limit of 10 %. For parameter μy, calculation of precision level is calculated as 8.1% for K2C, 5.6% for WB and 7.2% for VH which is used to remove outlier data for average wood fuel consumption for the sampled households. Both parameters are derived from survey with sample size accordance with the option (b) provided in the section 8.4 of the VMR0006 methodology/7/ which compliance with 90/10 confidence precision is not obligatory. Inconsistencies, such as sampling data variations and data discrepancies, identified between the MR, ER, and KPT have been addressed in findings CAR#2, CAR#3, CAR#4, and CAR#5. Each of these findings has been thoroughly reviewed and fully resolved to ensure alignment and data accuracy across all reports. The Sampling and surveys for CDM project activities and programmes of activities (Version 9.0)/31/ states under paragraph 28 that “When the project participants or the coordinating/managing entity have applied a sampling approach, the VVB may apply acceptance sampling as described in the steps indicated in paragraphs 29–38 below as part of validation/verification activities based on an AQL of 1 % and UQL of 15 %, producer risk 10 % and consumer risk 10 % which sample size of 25 and acceptance number of 1 is determined. The sampled users are taken from each project instance, 9 users from K2C, 8 from WB and 8 from VH.”
Excerpt read: 23,670 characters chosen by keyword, not the whole document
Evidence found for 2 of 3 checks (D1, D2); not found in the excerpt: D3.
“This is a new project and has not been registered or seeking registration under any other GHG programs.”
“This is a new project and has not been rejected by any other GHG program.”
“The project activity is not registered under any other GHG and non-GHG program or registry/12/. This has been confirmed by means of research on relevant applicable registries and available information in public domain and through declaration from the Project Owner.”
“The assessment team has searched for similar projects having same nature, capacity and project proponent as well as legal owner. The name of the owners mentions in the regulatory licenses are also matched and checked. It was concluded that no such projects having same location and geo-coordinates, technology or project/legal owners are registered in various carbon schemes like CDM, GCC, and Gold Standard.”
“5. The project is not active in any other crediting schemes other than the VCS. A signed declaration labelled “Double_Counting_Declaration_v1.0 signed (MP5)” is included as evidence.”
“(MP5).pdf” that the current project is not registered under any other GHG program.”
“The project is not included in an emissions trading program or any other mechanism that includes GHG allowance trading.”
“The project has not sought, nor received another form of GHG-related environmental credit.”
“No double claiming with emissions trading programs or binding emission limits”
“The project activity has been thoroughly search on the various websites of various emission trading programs and or ETS programs running in various counties, using nature, capacity and project proponent as well as legal owner as the search option. The project is not double claiming/35/ with other with other emission trading programs or binding emission limits. The assessment team has searched for similar projects having same nature, capacity and project proponent as well as legal owner. The name of the owners mentions in the regulatory licenses are also matched and checked. It was concluded that no such projects having same location and geo-coordinates, technology or project/legal owners are registered in other emission trading programs or binding emission limits various carbon schemes like CDM, GCC, and Gold Standard. Therefore, it is confirmed by VVB that the project is not double claiming with other emission trading programs or binding emission limits thereby complying with VCS Standard, v4.7; Sections 3.24.3.”
“No double claiming with other forms of environmental credit”
“As discussed above the, the project has been thoroughly search in the currently active environment schemes and found that the project is not registered with any other environmental credits. PP has also submitted a declaration/12/ stating that the project is not registered with any other registry (see CL#2). Therefore, it can be concluded that the project is not claiming any credits from other programs/schemes.”
“3. Section 1.10, MR: In order to avoid the double counting of the credits, PP shall provide a declaration stating the credits generated under this monitoring period will not be used with any other GHG programs and environmental credits”
“Are project reductions and removals or project activities also included in an emissions trading program or binding emission limit? See the VCS Program Definitions for definitions of emissions trading program and binding emission limit.”
“Has the project activity sought, received, or is planning to receive credit from another GHG- related environmental credit system? See the VCS Program Definitions for definition of GHG- related environmental credit system.”
“Are project reductions and removals or project activities also included in an emissions trading program or binding emission limit? See the VCS Program Definitions for definitions of emissions trading program and binding emission limit.”
“Has the project activity sought, received, or is planning to receive credit from another GHG-related environmental credit system? See the VCS Program Definitions for definition of GHG-related environmental credit system.”
“Are project reductions and removals or project activities also included in an emissions trading program or binding emission limit? See the VCS Program Definitions for definitions of emissions trading program and binding emission limit.”
“Has the project activity sought, received, or is planning to receive credit from another GHG- related environmental credit system? See the VCS Program Definitions for definition of GHG- related environmental credit system.”
Excerpt read: 43,360 characters chosen by keyword, not the whole document
Evidence found for 4 of 4 checks (S1, S2, S3, S4).
“i. Online meeting, 22/01/2021 ii. Webinar, 22/04/2021 iii. Sedawa Municipal Hall, 11/05/2021 iv. Lulekani Municipal Hall, 12/05/2021 v. Brooklyn Nazarene Church, 13/05/2021 vi. Mnisi Tribal Hall, 14/05/2021”
“Stakeholders have been provided with the contact details of K2C and TASC during the stakeholder consultations and at the point of distribution. Grievances may be reported telephonically, via email or in person to K2C or TASC, contact details of which are provided on all project documentation, including the non- technical summary and the ‘how to guide’ provided to all end users. Grievances will be addressed by the IP or the TASC PI manager and reported. These shall be dealt with as follows: - By responding to the stakeholder in”
“No Listed Activities requiring an environmental impact assessment under the National Environmental Management Act (NEMA, Act 107 of 1998), Environmental Impact”
“1) The project proponent shall recognize, respect and support local stakeholders’ property rights and where feasible, take measures to help secure rights. The project shall not encroach on private, stakeholder or government property or relocate people off their lands without consent. The project may affect property rights if free, prior and informed consent is obtained from those concerned and a”
Every version anchored on Sepolia stays readable: a retirement records the version current when it happened. Each file is hashed in your browser and compared with that version on the contract.
0xc459…b488(opens Sepolia Etherscan in a new tab)0xf5a4…a64a(opens Sepolia Etherscan in a new tab)0xba55…4388(opens Sepolia Etherscan in a new tab)0xfef1…19cf(opens Sepolia Etherscan in a new tab)0x5bc5…cba6(opens Sepolia Etherscan in a new tab)0x3964…2f98(opens Sepolia Etherscan in a new tab)Signal, not verdict. Each check says whether a specific kind of evidence is present in an excerpt of the submitted documents, not whether the evidence is true. Scores are not an ICVCM assessment, a rating of the project, or a statement about its real-world impact.
Report v6 anchor on Sepolia Etherscan(opens Sepolia Etherscan in a new tab)