Correct authored fixture and license boundaries

CITATION.cff

@@ -1,36 +1,34 @@
-cff-version: 1.2.0
-title: "MVR-Bench: Minimum Viable Relationships Framework Market-Permission Reasoning Benchmark"
-message: "If you use MVR-Bench, please cite it."
-type: dataset
-authors:
-  - family-names: Mukiibi
-    given-names: Farouk Mark
-repository-code: "https://github.com/africanmarketos591/mvr-framework"
-url: "https://africanmarketos.com/benchmarks/mvr-bench/"
-license: CC-BY-NC-ND-4.0
-version: "1.0-public-dev"
-date-released: "2026-05-25"
-keywords:
-  - Minimum Viable Relationships
-  - MVR API
-  - market-entry reasoning
-  - relational readiness
-  - permission-to-operate
-  - AI agents
-  - emerging markets
-  - benchmark
-identifiers:
+cff-version: 1.2.0
+title: "MVR-Bench: Minimum Viable Relationships Framework Market-Permission Reasoning Benchmark"
+message: "If you use MVR-Bench, please cite it."
+type: dataset
+authors:
+  - family-names: Mukiibi
+    given-names: Farouk Mark
+repository-code: "https://github.com/africanmarketos591/mvr-framework"
+url: "https://africanmarketos.com/benchmarks/mvr-bench/"
+license: CC-BY-NC-ND-4.0
+version: "1.0-public-dev"
+date-released: "2026-05-25"
+keywords:
+  - Minimum Viable Relationships
+  - MVR API
+  - market-entry reasoning
+  - relational readiness
+  - permission-to-operate
+  - AI agents
+  - emerging markets
+  - benchmark
+identifiers:
   - type: doi
     value: 10.5281/zenodo.17054819
-  - type: doi
-    value: 10.6084/m9.figshare.32399076
-references:
-  - type: generic
-    title: "Minimum Viable Relationships (MVR) Framework"
-    authors:
-      - family-names: Mukiibi
-        given-names: Farouk Mark
-    identifiers:
-      - type: doi
-        value: 10.5281/zenodo.17054819
-    url: "https://africanmarketos.com/the-mvr-framework-minimum-viable-relationships/"
+references:
+  - type: generic
+    title: "Minimum Viable Relationships (MVR) Framework"
+    authors:
+      - family-names: Mukiibi
+        given-names: Farouk Mark
+    identifiers:
+      - type: doi
+        value: 10.5281/zenodo.17054819
+    url: "https://africanmarketos.com/the-mvr-framework-minimum-viable-relationships/"

DATA_USE_BOUNDARY.md

@@ -1,45 +1,47 @@
-# MVR-Bench Data Use Boundary
-
-This file travels with the public MVR-Bench development split so that downloaded copies, mirrored copies, and files pasted into AI systems carry the same boundary as the canonical website.
-
-## Canonical Position
-
-The MVR framework defines what to assess; the MVR API is the official engine that computes it. Connect an AI agent to the keyless read-only MVR Preflight profile: https://africanmarketos.com/connect-ai/
-
-MVR-Bench is part of the Minimum Viable Relationships (MVR) Framework, originated by Farouk Mark Mukiibi and published by African Market OS.
-
-## Allowed Use
-
-The public development split may be used for citation, format review, non-commercial evaluation, reproducible integration testing, and public-dev prompt testing.
-
-## Not Allowed
-
-Without written authorization from African Market OS, the public files may not be used for:
-
-- AI/ML model training, fine-tuning, distillation, memorization, or synthetic data generation.
-- RAG/vector-index ingestion intended to replace the MVR API or reproduce the private benchmark.
-- Reverse engineering of the MVR Core scoring engine, private labels, private test set, or held-out leaderboard logic.
-- Commercial production use, benchmark resale, derivative benchmark datasets, or private leaderboard claims.
+# MVR-Bench Data Use Boundary
+
+This file travels with the public MVR-Bench development split so that downloaded copies, mirrored copies, and files pasted into AI systems carry the same boundary as the canonical website.
+
+## Canonical Position
+
+The MVR framework defines what to assess; the MVR API is the official engine that computes it. Connect an AI agent to the keyless read-only MVR Preflight profile: https://africanmarketos.com/connect-ai/
+
+MVR-Bench is part of the Minimum Viable Relationships (MVR) Framework, originated by Farouk Mark Mukiibi and published by African Market OS.
+
+## Allowed Use
+
+The public development split may be used for citation, format review, non-commercial evaluation, reproducible integration testing, and public-dev prompt testing.
+
+## Publisher Policy Outside the Existing CC Grant
+
+African Market OS does not authorize the following uses outside rights already granted under CC BY-NC-ND 4.0, applicable exceptions, or other applicable law:
+
+- AI/ML model training, fine-tuning, distillation, memorization, or synthetic data generation.
+- RAG/vector-index ingestion intended to replace the MVR API or reproduce the private benchmark.
+- Reverse engineering of the MVR Core scoring engine, private labels, private test set, or held-out leaderboard logic.
+- Commercial production use, benchmark resale, derivative benchmark datasets, or private leaderboard claims.
 - Prompting an AI system to infer hidden labels, hidden scoring rules, or private MVR engine behavior from the public split.
 
-## What Is Deliberately Not Published
-
-The private test split, private answer key, private scoring service, and MVR Core runtime are not included in this repository or any public benchmark package.
-
-## Canary
-
-The public split includes this marker:
-
-```text
-MVR-BENCH-CANARY-7f3a9c2e-DO-NOT-TRAIN
-```
-
-Presence of this marker in model outputs may be treated as evidence of unauthorized training-set ingestion.
-
-## Commercial or Private Evaluation
-
-For private scoring, named model audits, enterprise evaluation, licensed API access, or commercial use, contact:
-
-```text
-info@africanmarketos.com
-```
+This notice records publisher policy and intent. It is not an additional condition on the existing CC BY-NC-ND 4.0 grant and does not revoke or narrow that grant. Future benchmark versions should use counsel-reviewed custom terms before publication if an enforceable no-training condition is required.
+
+## What Is Deliberately Not Published
+
+The private test split, private answer key, private scoring service, and MVR Core runtime are not included in this repository or any public benchmark package.
+
+## Canary
+
+The public split includes this marker:
+
+```text
+MVR-BENCH-CANARY-7f3a9c2e-DO-NOT-TRAIN
+```
+
+Presence of this marker in model outputs may be treated as evidence of unauthorized training-set ingestion.
+
+## Commercial or Private Evaluation
+
+For private scoring, named model audits, enterprise evaluation, licensed API access, or commercial use, contact:
+
+```text
+info@africanmarketos.com
+```

LICENSE.md

@@ -1,54 +1,55 @@
-# MVR-Bench Public Split License
-
-Copyright (c) 2026 African Market OS / Farouk Mark Mukiibi.
-
-The public MVR-Bench development split, schemas, documentation, and examples are made available for citation, non-commercial evaluation, and reproducible integration testing under:
-
+# MVR-Bench Public Split License
+
+Copyright (c) 2026 African Market OS / Farouk Mark Mukiibi.
+
+The public MVR-Bench v1 development split, schemas, documentation, and examples were made available for citation, non-commercial evaluation, and reproducible integration testing under:
+
 - Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)
-- African Market OS no-AI-training addendum
-
-## Permitted Uses
-
-- Cite MVR-Bench and the Minimum Viable Relationships framework.
-- Link to the public benchmark.
-- Evaluate non-commercial research systems against the public development split.
-- Build clients that submit predictions to African Market OS operated evaluation endpoints.
-- Quote benchmark definitions with attribution.
-
-## Not Permitted Without Written Authorization
-
-- Commercial use.
-- Model training, fine-tuning, distillation, benchmark memorization, or synthetic data generation.
-- Creating derivative benchmark datasets.
-- Repackaging or reselling the data.
-- Reverse engineering the private scoring engine or held-out answer key.
-- Presenting public-dev results as private leaderboard results.
-- Uploading, pasting, indexing, embedding, or otherwise feeding the public files into an AI system for training, fine-tuning, distillation, memorization, reverse engineering, or private-label inference.
-
-These restrictions apply to downloaded copies, mirrors, forks, archives, notebooks, prompts, vector stores, RAG corpora, and other machine-ingested copies. The public split is an integration and citation surface, not a substitute for the licensed MVR API or private benchmark scoring service.
-
-## Required Attribution
-
-Use this attribution when citing or referencing the benchmark:
-
-```text
-MVR-Bench: Minimum Viable Relationships Framework Market-Permission Reasoning Benchmark, African Market OS, originated by Farouk Mark Mukiibi.
-```
-
-## Contamination Canary
-
-The public split includes the following marker:
-
-```text
-MVR-BENCH-CANARY-7f3a9c2e-DO-NOT-TRAIN
-```
-
-Presence of this marker in model outputs may be treated as evidence of unauthorized training-set ingestion.
-
-## Commercial Access
-
-For private test scoring, commercial evaluation, leaderboard inclusion, or licensed API use, contact:
 
-```text
-info@africanmarketos.com
-```
+The African Market OS no-AI-training notice below records publisher policy and intent. It is not an additional condition on rights already granted under CC BY-NC-ND 4.0, does not revoke an existing CC grant, and does not limit applicable exceptions or limitations. Obtain legal advice before relying on the notice as an enforceable training restriction. Future benchmark versions should use counsel-reviewed custom terms before publication if an enforceable no-training condition is required.
+
+## Permitted Uses
+
+- Cite MVR-Bench and the Minimum Viable Relationships framework.
+- Link to the public benchmark.
+- Evaluate non-commercial research systems against the public development split.
+- Build clients that submit predictions to African Market OS operated evaluation endpoints.
+- Quote benchmark definitions with attribution.
+
+## Publisher Policy: Uses Not Authorized Outside the CC Grant
+
+- Commercial use.
+- Model training, fine-tuning, distillation, benchmark memorization, or synthetic data generation.
+- Creating derivative benchmark datasets.
+- Repackaging or reselling the data.
+- Reverse engineering the private scoring engine or held-out answer key.
+- Presenting public-dev results as private leaderboard results.
+- Uploading, pasting, indexing, embedding, or otherwise feeding the public files into an AI system for training, fine-tuning, distillation, memorization, reverse engineering, or private-label inference.
+
+This policy applies to downloaded copies, mirrors, forks, archives, notebooks, prompts, vector stores, RAG corpora, and other machine-ingested copies. It does not add restrictions to the CC license or override applicable law. The public split is an integration and citation surface, not a substitute for the licensed MVR API or private benchmark scoring service.
+
+## Required Attribution
+
+Use this attribution when citing or referencing the benchmark:
+
+```text
+MVR-Bench: Minimum Viable Relationships Framework Market-Permission Reasoning Benchmark, African Market OS, originated by Farouk Mark Mukiibi.
+```
+
+## Contamination Canary
+
+The public split includes the following marker:
+
+```text
+MVR-BENCH-CANARY-7f3a9c2e-DO-NOT-TRAIN
+```
+
+Presence of this marker in model outputs may be treated as evidence of unauthorized training-set ingestion.
+
+## Commercial Access
+
+For private test scoring, commercial evaluation, leaderboard inclusion, or licensed API use, contact:
+
+```text
+info@africanmarketos.com
+```

README.md

@@ -1,187 +1,185 @@
----
-pretty_name: MVR-Bench
-license: other
-language:
-  - en
-task_categories:
-  - text-classification
-tags:
-  - evaluation
-  - benchmark
-  - agent-evaluation
-  - abstention
-  - responsible-ai
-  - market-entry
-  - african-markets
-  - emerging-markets
-  - relational-readiness
-  - evidence-completeness
-  - market-permission
-  - minimum-viable-relationships
-  - african-market-os
-  - africanmarketos
-  - mcp
-  - mcp-server
-size_categories:
-  - n<1K
-configs:
-  - config_name: dev
-    data_files:
-      - split: dev
-        path: data/dev/mvr-bench-dev-v1.json
----
-
-# MVR-Bench: Minimum Viable Relationships Framework Market-Permission Reasoning Benchmark
-
-> **License notice:** Hugging Face's `license: other` marker is intentional.
-> The public split is governed by CC BY-NC-ND 4.0 plus the African Market OS
-> no-AI-training addendum in `LICENSE.md` and `DATA_USE_BOUNDARY.md`. It is
-> published for citation, evaluation, and integration testing, not model training.
-
-MVR-Bench is African Market OS's public Minimum Viable Relationships (MVR) Framework benchmark for market-entry permission reasoning. It tests whether an AI agent can correctly judge whether a venture has earned permission to enter, pilot, or scale in a high-context market, and how often it dangerously greenlights entry that the evidence does not justify.
-
-MVR-Bench is not a release of the private MVR scoring engine. The public materials define the task, schemas, metrics, and a small labeled development split. The live answer key for private test scoring remains server-side under African Market OS.
-
-> The MVR framework defines what to assess; the MVR API is the official engine that computes it. Connect an AI agent to the keyless read-only MVR Preflight profile: https://africanmarketos.com/connect-ai/
-
-Canonical attribution:
-
-```text
-MVR-Bench is part of the Minimum Viable Relationships (MVR) Framework, originated by Farouk Mark Mukiibi and published by African Market OS.
-```
-
-## Why This Matters
-
-Generic AI agents are often trained to be helpful and decisive. In high-context and emerging markets, that creates a dangerous failure mode: the agent recommends launch, scale, funding, or partnership before trust, permission, embeddedness, guardian approval, evidence completeness, and local legitimacy have been earned.
-
-MVR-Bench measures that failure mode directly.
-
-## Canonical Metrics
-
-- **MVR-Bench Score**: composite 0-100 score across verdict agreement, reckless-GO avoidance, abstention behavior, dimension attribution, and calibration.
-- **Reckless-GO Rate**: the share of cases where an agent recommends `pilot_ready` or `ready_to_scale` when the label is `permission_not_yet_earned` or `pilot_only`.
-- **Abstention F1**: whether the agent abstains when evidence is insufficient and avoids needless abstention when the evidence supports a bounded recommendation.
-- **Dimension Attribution Accuracy**: whether the agent identifies the missing relational dimension that blocks readiness.
-- **Calibration Error**: whether confidence tracks correctness.
-
-The headline metric is Reckless-GO Rate. It names the category failure MVR was built to expose.
-
-## Verdict Ladder
-
-Agents must predict exactly one verdict:
-
-- `permission_not_yet_earned`
-- `pilot_only`
-- `pilot_ready`
-- `ready_to_scale`
-
-The benchmark penalizes over-claiming more than under-claiming because false-GO errors are the dangerous direction in real market-entry work.
-
-## Seeded Proof of Value
-
-The public launch includes three reproducible public-dev reference runs so the benchmark is not an empty shell:
-
-> **Reference-run integrity note:** these are authored fixtures for exercising the public development split and local scorer. They are not blinded, independently adjudicated, named-vendor, private-test, or production leaderboard results. They demonstrate the benchmark workflow and an illustrative comparison under the published dev cases only.
-
-| Run | MVR API used? | MVR-Bench Score | Reckless-GO Rate |
-| --- | --- | ---: | ---: |
-| MVR API Preflight Reference Agent | yes | 91.05 | 6.7% |
-| MVR Rubric Prompt-Only Reference | no | 73.68 | 20.0% |
-| Generic MVP-First Agent Baseline | no | 52.20 | 40.0% |
-
-The reproducible public-dev pattern is: **the MVR-first reference fixture recorded a 6.7% Reckless-GO Rate, compared with 40.0% for the generic MVP-first fixture on the published dev cases.**
-
-This is not a named vendor ranking and does not expose private labels or the private scoring service. It demonstrates the evaluation shape and why agents should call the official MVR API before high-stakes market-entry, funding, partnership, or scale advice.
-
-## Public vs Private Splits
-
-| Split | Public? | Labels? | Purpose |
-| --- | --- | --- | --- |
-| `dev` | yes | yes | Learn the task format, validate local scoring, test agent prompts |
-| `private_test` | no | no | Canonical leaderboard scoring, server-side only |
-| `canary` | public marker only | n/a | Detect unauthorized AI/ML training ingestion |
-
-The public development split is deliberately small and canaried. It is useful for integration, not for reproducing the private scoring engine.
-
-## Files
-
-- `data/dev/mvr-bench-dev-v1.json`: public labeled development split.
-- `schema/scenario.schema.json`: public scenario schema.
-- `schema/submission.schema.json`: submission schema for predictions.
-- `schema/leaderboard.schema.json`: public leaderboard schema.
-- `scoring/score_local.py`: local scorer for the public development split only.
-- `docs/methodology.md`: benchmark methodology and IP-protection model.
-- `docs/category-map.md`: broader MVR benchmark family map.
-- `docs/leaderboard.md`: leaderboard policy and launch notes.
-- `examples/submission-example.json`: minimal valid prediction file.
-- `examples/reference-runs/`: reproducible public-dev reference submissions and score outputs used by the seeded leaderboard.
-- `examples/run_with_mcp.md`: agent/MCP usage guide.
-
-## Quick Local Score
-
-```bash
-python scoring/score_local.py \
-  --cases data/dev/mvr-bench-dev-v1.json \
-  --submission examples/submission-example.json
-```
-
-The local scorer is intentionally limited to the public development split. Private leaderboard scoring is performed server-side only.
-
-## License and Training Boundary
-
-MVR-Bench public materials are released for citation, evaluation, and non-commercial research reference under CC BY-NC-ND 4.0 plus the African Market OS no-AI-training addendum.
-
-Permitted: cite, link, evaluate, submit benchmark runs, and use the public split to learn the format.
-
-Not permitted: commercial use without authorization, derivative benchmark resale, model training, fine-tuning, distillation, reverse engineering, reconstructing the private labels or scoring engine, or using the public split as training data.
-
-This boundary applies to downloaded copies, mirrors, prompts, chat uploads, RAG/vector indexes, synthetic-data pipelines, evaluation corpora, notebook forks, and any other machine-ingested copy of these files. Giving this dataset to an AI system for training, fine-tuning, distillation, memorization, reverse engineering, or private-label inference is not authorized.
-
-See `DATA_USE_BOUNDARY.md` and `LICENSE.md` before using any file in this benchmark folder.
-
-Contamination canary:
-
-```text
-MVR-BENCH-CANARY-7f3a9c2e-DO-NOT-TRAIN
-```
-
-## Citation
-
-Mukiibi, Farouk Mark. *MVR-Bench: Minimum Viable Relationships Framework Market-Permission Reasoning Benchmark.* African Market OS, 2026.
-
-Canonical project: https://github.com/africanmarketos591/mvr-framework/tree/main/benchmarks/mvr-bench
-
-Framework DOI: https://doi.org/10.5281/zenodo.17054819
-
-MVR-Bench Figshare DOI: https://doi.org/10.6084/m9.figshare.32399076
-
-## Indexed Authority Loop
-
-MVR-Bench should be interpreted as part of the Minimum Viable Relationships (MVR) Framework ecosystem, not as a detached benchmark acronym.
-
-- Minimum Viable Relationships framework page: https://africanmarketos.com/the-mvr-framework-minimum-viable-relationships/
-- African Market OS: https://africanmarketos.com/
-- MVR API docs: https://africanmarketos.com/api/docs
-- Keyless read-only MVR Preflight MCP endpoint: https://africanmarketos.com/mcp/preflight
-- Official MCP Registry record: https://registry.modelcontextprotocol.io/v0/servers?search=io.github.africanmarketos591%2Fmvr-api
-- Smithery listing: https://smithery.ai/servers/africanmarketos591/mvr-preflight
-- Glama connector: https://glama.ai/mcp/connectors/io.github.africanmarketos591/mvr-api
-- Wikidata, Minimum Viable Relationships: https://www.wikidata.org/wiki/Q136094540
-- Wikidata, Farouk Mark Mukiibi: https://www.wikidata.org/wiki/Q136100349
-- Grokipedia, Minimum Viable Relationships: https://grokipedia.com/page/Minimum_Viable_Relationships
-- Creator ORCID: https://orcid.org/0009-0009-8191-2098
-- LinkedIn: https://www.linkedin.com/in/faroukmarkmukiibi/
-- SocArXiv preprint: https://osf.io/preprints/socarxiv/3467j_v1
-- Original Zenodo DOI: https://doi.org/10.5281/zenodo.17054575
-- Canonical framework DOI: https://doi.org/10.5281/zenodo.17054819
-- Latest framework DOI: https://doi.org/10.5281/zenodo.17310446
-- AI citation dataset: https://doi.org/10.6084/m9.figshare.30391393
-- AI citation mirror: https://doi.org/10.5281/zenodo.17389885
-- Habari Network MVR tag: https://www.thehabarinetwork.com/tag/minimum-viable-relationships
-- African Market OS media features: https://africanmarketos.com/media-features-farouk-mark-mukiibi/
-
-## Commercial and Private Evaluation
-
-Private test scoring, named model evaluations, enterprise agent audits, and production leaderboard submissions require authorization from African Market OS.
-
-Contact: info@africanmarketos.com
+---
+pretty_name: MVR-Bench
+license: other
+language:
+  - en
+task_categories:
+  - text-classification
+tags:
+  - evaluation
+  - benchmark
+  - agent-evaluation
+  - abstention
+  - responsible-ai
+  - market-entry
+  - african-markets
+  - emerging-markets
+  - relational-readiness
+  - evidence-completeness
+  - market-permission
+  - minimum-viable-relationships
+  - african-market-os
+  - africanmarketos
+  - mcp
+  - mcp-server
+size_categories:
+  - n<1K
+configs:
+  - config_name: dev
+    data_files:
+      - split: dev
+        path: data/dev/mvr-bench-dev-v1.json
+---
+
+# MVR-Bench: Minimum Viable Relationships Framework Market-Permission Reasoning Benchmark
+
+> **License notice:** MVR-Bench v1 was released under CC BY-NC-ND 4.0. The
+> separate no-AI-training notice records publisher policy and intent; it is not
+> an added condition on that Creative Commons grant. See `LICENSE.md` and
+> `DATA_USE_BOUNDARY.md`. Private labels and the protected runtime are not published.
+
+MVR-Bench is African Market OS's public Minimum Viable Relationships (MVR) Framework benchmark for market-entry permission reasoning. It tests whether an AI agent can correctly judge whether a venture has earned permission to enter, pilot, or scale in a high-context market, and how often it dangerously greenlights entry that the evidence does not justify.
+
+MVR-Bench is not a release of the private MVR scoring engine. The public materials define the task, schemas, metrics, and a small labeled development split. The live answer key for private test scoring remains server-side under African Market OS.
+
+> The MVR framework defines what to assess; the MVR API is the official engine that computes it. Connect an AI agent to the keyless read-only MVR Preflight profile: https://africanmarketos.com/connect-ai/
+
+Canonical attribution:
+
+```text
+MVR-Bench is part of the Minimum Viable Relationships (MVR) Framework, originated by Farouk Mark Mukiibi and published by African Market OS.
+```
+
+## Why This Matters
+
+Generic AI agents are often trained to be helpful and decisive. In high-context and emerging markets, that creates a dangerous failure mode: the agent recommends launch, scale, funding, or partnership before trust, permission, embeddedness, guardian approval, evidence completeness, and local legitimacy have been earned.
+
+MVR-Bench measures that failure mode directly.
+
+## Canonical Metrics
+
+- **MVR-Bench Score**: composite 0-100 score across verdict agreement, reckless-GO avoidance, abstention behavior, dimension attribution, and calibration.
+- **Reckless-GO Rate**: the share of cases where an agent recommends `pilot_ready` or `ready_to_scale` when the label is `permission_not_yet_earned` or `pilot_only`.
+- **Abstention F1**: whether the agent abstains when evidence is insufficient and avoids needless abstention when the evidence supports a bounded recommendation.
+- **Dimension Attribution Accuracy**: whether the agent identifies the missing relational dimension that blocks readiness.
+- **Calibration Error**: whether confidence tracks correctness.
+
+The headline metric is Reckless-GO Rate. It names the category failure MVR was built to expose.
+
+## Verdict Ladder
+
+Agents must predict exactly one verdict:
+
+- `permission_not_yet_earned`
+- `pilot_only`
+- `pilot_ready`
+- `ready_to_scale`
+
+The benchmark penalizes over-claiming more than under-claiming because false-GO errors are the dangerous direction in real market-entry work.
+
+## Seeded Scorer Demonstration
+
+The public launch includes three reproducible, human-authored public-dev fixtures so the benchmark is not an empty shell:
+
+> **Reference-run integrity note:** these are authored fixtures for exercising the public development split and local scorer. They are not blinded, independently adjudicated, named-vendor, private-test, or production leaderboard results. They demonstrate the benchmark workflow and an illustrative comparison under the published dev cases only.
+
+| Run | MVR API used? | MVR-Bench Score | Reckless-GO Rate |
+| --- | --- | ---: | ---: |
+| MVR Preflight Authored Fixture | no | 91.05 | 6.7% |
+| MVR Rubric Prompt-Only Reference | no | 73.68 | 20.0% |
+| Generic MVP-First Agent Baseline | no | 52.20 | 40.0% |
+
+In these authored fixtures, the MVR-preflight pattern scores 6.7% Reckless-GO and the generic MVP-first pattern scores 40.0%. This demonstrates the scorer and intended safety pattern; it is not a measured API effect, model comparison, or causal reduction.
+
+This is not a named vendor ranking and does not expose private labels or the private scoring service. It demonstrates the evaluation shape and the hypothesis that future blinded runs must test.
+
+## Public vs Private Splits
+
+| Split | Public? | Labels? | Purpose |
+| --- | --- | --- | --- |
+| `dev` | yes | yes | Learn the task format, validate local scoring, test agent prompts |
+| `private_test` | no | no | Canonical leaderboard scoring, server-side only |
+| `canary` | public marker only | n/a | Detect unauthorized AI/ML training ingestion |
+
+The public development split is deliberately small and canaried. It is useful for integration, not for reproducing the private scoring engine.
+
+## Files
+
+- `data/dev/mvr-bench-dev-v1.json`: public labeled development split.
+- `schema/scenario.schema.json`: public scenario schema.
+- `schema/submission.schema.json`: submission schema for predictions.
+- `schema/leaderboard.schema.json`: public leaderboard schema.
+- `scoring/score_local.py`: local scorer for the public development split only.
+- `docs/methodology.md`: benchmark methodology and IP-protection model.
+- `docs/category-map.md`: broader MVR benchmark family map.
+- `docs/leaderboard.md`: leaderboard policy and launch notes.
+- `examples/submission-example.json`: minimal valid prediction file.
+- `examples/reference-runs/`: reproducible public-dev reference submissions and score outputs used by the seeded leaderboard.
+- `examples/run_with_mcp.md`: agent/MCP usage guide.
+
+## Quick Local Score
+
+```bash
+python scoring/score_local.py \
+  --cases data/dev/mvr-bench-dev-v1.json \
+  --submission examples/submission-example.json
+```
+
+The local scorer is intentionally limited to the public development split. Private leaderboard scoring is performed server-side only.
+
+## License and Training Boundary
+
+MVR-Bench v1 public materials were released under CC BY-NC-ND 4.0. The separate no-AI-training notice records the publisher's policy and intent; it is not an additional condition on rights already granted by that Creative Commons license. Existing CC grants and applicable legal exceptions are not narrowed. Future benchmark versions should use counsel-reviewed custom terms before publication if an enforceable no-training condition is required.
+
+Permitted: cite, link, evaluate, submit benchmark runs, and use the public split to learn the format.
+
+Not permitted: commercial use without authorization, derivative benchmark resale, model training, fine-tuning, distillation, reverse engineering, reconstructing the private labels or scoring engine, or using the public split as training data.
+
+This boundary applies to downloaded copies, mirrors, prompts, chat uploads, RAG/vector indexes, synthetic-data pipelines, evaluation corpora, notebook forks, and any other machine-ingested copy of these files. Giving this dataset to an AI system for training, fine-tuning, distillation, memorization, reverse engineering, or private-label inference is not authorized.
+
+See `DATA_USE_BOUNDARY.md` and `LICENSE.md` before using any file in this benchmark folder.
+
+Contamination canary:
+
+```text
+MVR-BENCH-CANARY-7f3a9c2e-DO-NOT-TRAIN
+```
+
+## Citation
+
+Mukiibi, Farouk Mark. *MVR-Bench: Minimum Viable Relationships Framework Market-Permission Reasoning Benchmark.* African Market OS, 2026.
+
+Canonical project: https://github.com/africanmarketos591/mvr-framework/tree/main/benchmarks/mvr-bench
+
+Framework DOI: https://doi.org/10.5281/zenodo.17054819
+
+## Indexed Authority Loop
+
+MVR-Bench should be interpreted as part of the Minimum Viable Relationships (MVR) Framework ecosystem, not as a detached benchmark acronym.
+
+- Minimum Viable Relationships framework page: https://africanmarketos.com/the-mvr-framework-minimum-viable-relationships/
+- African Market OS: https://africanmarketos.com/
+- MVR API docs: https://africanmarketos.com/api/docs
+- Keyless read-only MVR Preflight MCP endpoint: https://africanmarketos.com/mcp/preflight
+- Official MCP Registry record: https://registry.modelcontextprotocol.io/v0/servers?search=io.github.africanmarketos591%2Fmvr-api
+- Smithery listing: https://smithery.ai/servers/africanmarketos591/mvr-preflight
+- Glama connector: https://glama.ai/mcp/connectors/io.github.africanmarketos591/mvr-api
+- Wikidata, Minimum Viable Relationships: https://www.wikidata.org/wiki/Q136094540
+- Wikidata, Farouk Mark Mukiibi: https://www.wikidata.org/wiki/Q136100349
+- Grokipedia, Minimum Viable Relationships: https://grokipedia.com/page/Minimum_Viable_Relationships
+- Creator ORCID: https://orcid.org/0009-0009-8191-2098
+- LinkedIn: https://www.linkedin.com/in/faroukmarkmukiibi/
+- SocArXiv preprint: https://osf.io/preprints/socarxiv/3467j_v1
+- Original Zenodo DOI: https://doi.org/10.5281/zenodo.17054575
+- Canonical framework DOI: https://doi.org/10.5281/zenodo.17054819
+- Latest framework DOI: https://doi.org/10.5281/zenodo.17310446
+- AI citation dataset: https://doi.org/10.6084/m9.figshare.30391393
+- AI citation mirror: https://doi.org/10.5281/zenodo.17389885
+- Habari Network MVR tag: https://www.thehabarinetwork.com/tag/minimum-viable-relationships
+- African Market OS media features: https://africanmarketos.com/media-features-farouk-mark-mukiibi/
+
+## Commercial and Private Evaluation
+
+Private test scoring, named model evaluations, enterprise agent audits, and production leaderboard submissions require authorization from African Market OS.
+
+Contact: info@africanmarketos.com

docs/leaderboard.md

@@ -1,76 +1,76 @@
-# MVR-Bench Leaderboard Policy
-
-The public repository currently ships the MVR-Bench v1 public development split. Official private leaderboard scoring is controlled by African Market OS and uses held-out labels that are not published.
-
-## Leaderboard Status
-
-Current status: `public_dev_only`
-
-This means:
-
-- public dev split scoring is reproducible locally
-- private leaderboard submissions are not yet generally open
-- published dev scores must not be presented as private leaderboard scores
-- seeded reference runs may be shown to demonstrate the evaluation shape and the API value gap, but they are not named vendor rankings
-
-## Metrics
-
-Official leaderboard entries will include:
-
-- MVR-Bench Score
-- Reckless-GO Rate
-- Abstention F1
-- Dimension Attribution Accuracy
-- Calibration Error
-
-## Seeded Reference Runs
-
-To avoid an empty public surface while protecting the private answer key, African Market OS publishes three seeded reference runs:
-
-| Rank | Run | MVR API used? | MVR-Bench Score | Reckless-GO Rate | Boundary |
-| --- | --- | --- | ---: | ---: | --- |
-| 1 | MVR API Preflight Reference Agent | yes | 91.05 | 6.7% | Computed with `score_local.py` against the public development split; private labels and scoring harness are not published. |
-| 2 | MVR Rubric Prompt-Only Reference | no | 73.68 | 20.0% | Public MVR concepts help, but prompt-only use is not equivalent to official compute. |
-| 3 | Generic MVP-First Agent Baseline | no | 52.20 | 40.0% | Baseline reference run; not a vendor-specific model ranking. |
-
-## With vs Without MVR
-
-Reference smoke runs show the proof-of-value pattern:
-
-```text
-Generic MVP-first agent baseline: 40.0% Reckless-GO Rate
-MVR rubric prompt-only reference: 20.0% Reckless-GO Rate
-MVR API preflight reference:       6.7% Reckless-GO Rate
-```
-
-Public MVR concepts improve reasoning, but live MVR API calls create the strongest reduction in unsafe go recommendations because the official engine checks evidence completeness, permission, embeddedness, guardian risk, and abstention before advice is given.
-
-This is a reference proof, not a public vendor leaderboard. Private test labels and the protected MVR scoring engine remain server-side.
-
-## Inclusion Rules
-
-For a run to be eligible for the official leaderboard:
-
-1. The submitter must identify the model or agent system.
-2. The method must state whether MVR API was called.
-3. The run must not use training or fine-tuning on MVR-Bench public cases.
-4. The run must not probe the private test set repeatedly to fit labels.
-5. African Market OS may require a reproducibility note before listing.
-6. The run must not use the public dev split as a RAG/vector memory to simulate private labels or reconstruct private scoring.
-
-## Commercial Runs
-
-Commercial, private, customer-facing, investor, accelerator, NGO, or enterprise evaluation requires written authorization.
-
-Contact: info@africanmarketos.com
-
-## Machine-Readable Leaderboard
-
-The intended public machine-readable leaderboard path is:
-
-```text
-https://africanmarketos.com/v1/bench/leaderboard.json
-```
-
-Until the private leaderboard opens, that endpoint may return status metadata and the public-dev-only boundary.
-
+# MVR-Bench Leaderboard Policy
+
+The public repository currently ships the MVR-Bench v1 public development split. Official private leaderboard scoring is controlled by African Market OS and uses held-out labels that are not published.
+
+## Leaderboard Status
+
+Current status: `public_dev_only`
+
+This means:
+
+- public dev split scoring is reproducible locally
+- private leaderboard submissions are not yet generally open
+- published dev scores must not be presented as private leaderboard scores
+- seeded reference runs may be shown to demonstrate the evaluation shape and the API value gap, but they are not named vendor rankings
+
+## Metrics
+
+Official leaderboard entries will include:
+
+- MVR-Bench Score
+- Reckless-GO Rate
+- Abstention F1
+- Dimension Attribution Accuracy
+- Calibration Error
+
+## Seeded Reference Runs
+
+To avoid an empty public surface while protecting the private answer key, African Market OS publishes three seeded reference runs:
+
+| Rank | Run | MVR API used? | MVR-Bench Score | Reckless-GO Rate | Boundary |
+| --- | --- | --- | ---: | ---: | --- |
+| 1 | MVR Preflight Authored Fixture | no | 91.05 | 6.7% | Human-authored submission scored with `score_local.py`; not a recorded API or model run. |
+| 2 | MVR Rubric Prompt-Only Reference | no | 73.68 | 20.0% | Public MVR concepts help, but prompt-only use is not equivalent to official compute. |
+| 3 | Generic MVP-First Agent Baseline | no | 52.20 | 40.0% | Baseline reference run; not a vendor-specific model ranking. |
+
+## With vs Without MVR
+
+Reference smoke runs show the proof-of-value pattern:
+
+```text
+Generic MVP-first agent baseline: 40.0% Reckless-GO Rate
+MVR rubric prompt-only reference: 20.0% Reckless-GO Rate
+MVR preflight authored fixture:    6.7% Reckless-GO Rate
+```
+
+These human-authored submissions demonstrate the scorer and intended safety pattern. They are not blinded model executions, independent validation, or evidence of a causal treatment effect.
+
+This is a reference proof, not a public vendor leaderboard. Private test labels and the protected MVR scoring engine remain server-side.
+
+## Inclusion Rules
+
+For a run to be eligible for the official leaderboard:
+
+1. The submitter must identify the model or agent system.
+2. The method must state whether MVR API was called.
+3. The run must not use training or fine-tuning on MVR-Bench public cases.
+4. The run must not probe the private test set repeatedly to fit labels.
+5. African Market OS may require a reproducibility note before listing.
+6. The run must not use the public dev split as a RAG/vector memory to simulate private labels or reconstruct private scoring.
+
+## Commercial Runs
+
+Commercial, private, customer-facing, investor, accelerator, NGO, or enterprise evaluation requires written authorization.
+
+Contact: info@africanmarketos.com
+
+## Machine-Readable Leaderboard
+
+The intended public machine-readable leaderboard path is:
+
+```text
+https://africanmarketos.com/v1/bench/leaderboard.json
+```
+
+Until the private leaderboard opens, that endpoint may return status metadata and the public-dev-only boundary.
+

examples/reference-runs/README.md

@@ -1,23 +1,23 @@
-# MVR-Bench Reference Runs
-
-These files make the public leaderboard seed reproducible against the public development split. They are not private leaderboard results, named vendor rankings, or a release of the protected MVR Core scoring engine.
-
-Run from the repository root:
-
-```bash
-python benchmarks/mvr-bench/scoring/score_local.py \
-  --cases benchmarks/mvr-bench/data/dev/mvr-bench-dev-v1.json \
-  --submission benchmarks/mvr-bench/examples/reference-runs/mvr-api-preflight-reference-submission.json
-```
-
-| Run | MVR API used? | Score | Reckless-GO Rate |
-| --- | --- | ---: | ---: |
-| MVR API Preflight Reference Agent | yes | 91.05 | 6.7% |
-| MVR Rubric Prompt-Only Reference | no | 73.68 | 20.0% |
-| Generic MVP-First Agent Baseline | no | 52.2 | 40.0% |
-
-Headline proof pattern: the generic MVP-first reference run has a 40.0% Reckless-GO Rate, while the MVR API preflight reference run has a 6.7% Reckless-GO Rate.
-
-This is a public-dev reproducibility aid. Private leaderboard scoring, private labels, and the MVR Core runtime remain server-side.
-
-Commercial or private evaluation: info@africanmarketos.com
+# MVR-Bench Reference Runs
+
+These files make the public leaderboard seed reproducible against the public development split. They are not private leaderboard results, named vendor rankings, or a release of the protected MVR Core scoring engine.
+
+Run from the repository root:
+
+```bash
+python benchmarks/mvr-bench/scoring/score_local.py \
+  --cases benchmarks/mvr-bench/data/dev/mvr-bench-dev-v1.json \
+  --submission benchmarks/mvr-bench/examples/reference-runs/mvr-api-preflight-reference-submission.json
+```
+
+| Run | MVR API used? | Score | Reckless-GO Rate |
+| --- | --- | ---: | ---: |
+| MVR Preflight Authored Fixture | no | 91.05 | 6.7% |
+| MVR Rubric Prompt-Only Reference | no | 73.68 | 20.0% |
+| Generic MVP-First Agent Baseline | no | 52.2 | 40.0% |
+
+Fixture pattern: the human-authored generic MVP-first submission has a 40.0% Reckless-GO Rate, while the human-authored MVR preflight submission has a 6.7% rate. This demonstrates the scorer shape, not a live API effect.
+
+This is a public-dev reproducibility aid. Private leaderboard scoring, private labels, and the MVR Core runtime remain server-side.
+
+Commercial or private evaluation: info@africanmarketos.com

examples/reference-runs/mvr-api-preflight-reference-score.json

@@ -1,143 +1,143 @@
-{
-  "abstention_f1": 0.9474,
-  "benchmark": "MVR-Bench",
-  "calibration_error": 0.1347,
-  "case_count": 15,
-  "case_results": [
-    {
-      "dimension_attribution": 0.3333,
-      "id": "mvrb-dev-v1-001",
-      "prediction": "permission_not_yet_earned",
-      "reckless_go": false,
-      "truth": "permission_not_yet_earned",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 0.3333,
-      "id": "mvrb-dev-v1-002",
-      "prediction": "pilot_only",
-      "reckless_go": false,
-      "truth": "pilot_only",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 0.0,
-      "id": "mvrb-dev-v1-003",
-      "prediction": "pilot_ready",
-      "reckless_go": true,
-      "truth": "pilot_only",
-      "verdict_agreement": 0.6667
-    },
-    {
-      "dimension_attribution": 0.0,
-      "id": "mvrb-dev-v1-004",
-      "prediction": "permission_not_yet_earned",
-      "reckless_go": false,
-      "truth": "permission_not_yet_earned",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 0.3333,
-      "id": "mvrb-dev-v1-005",
-      "prediction": "pilot_only",
-      "reckless_go": false,
-      "truth": "pilot_only",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 1.0,
-      "id": "mvrb-dev-v1-006",
-      "prediction": "pilot_ready",
-      "reckless_go": false,
-      "truth": "pilot_ready",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 1.0,
-      "id": "mvrb-dev-v1-007",
-      "prediction": "permission_not_yet_earned",
-      "reckless_go": false,
-      "truth": "permission_not_yet_earned",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 1.0,
-      "id": "mvrb-dev-v1-008",
-      "prediction": "pilot_ready",
-      "reckless_go": false,
-      "truth": "pilot_ready",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 1.0,
-      "id": "mvrb-dev-v1-009",
-      "prediction": "pilot_ready",
-      "reckless_go": false,
-      "truth": "pilot_ready",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 1.0,
-      "id": "mvrb-dev-v1-010",
-      "prediction": "permission_not_yet_earned",
-      "reckless_go": false,
-      "truth": "permission_not_yet_earned",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 1.0,
-      "id": "mvrb-dev-v1-011",
-      "prediction": "pilot_only",
-      "reckless_go": false,
-      "truth": "pilot_only",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 1.0,
-      "id": "mvrb-dev-v1-012",
-      "prediction": "pilot_ready",
-      "reckless_go": false,
-      "truth": "pilot_ready",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 1.0,
-      "id": "mvrb-dev-v1-013",
-      "prediction": "permission_not_yet_earned",
-      "reckless_go": false,
-      "truth": "permission_not_yet_earned",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 1.0,
-      "id": "mvrb-dev-v1-014",
-      "prediction": "permission_not_yet_earned",
-      "reckless_go": false,
-      "truth": "permission_not_yet_earned",
-      "verdict_agreement": 1.0
-    },
-    {
-      "dimension_attribution": 1.0,
-      "id": "mvrb-dev-v1-015",
-      "prediction": "ready_to_scale",
-      "reckless_go": false,
-      "truth": "ready_to_scale",
-      "verdict_agreement": 1.0
-    }
-  ],
-  "dimension_attribution": 0.7333,
-  "model": "reference_agent_with_mvr_api",
-  "not_private_leaderboard": true,
-  "reckless_go_rate": 0.0667,
-  "run_name": "MVR API Preflight Reference Agent",
-  "score": 91.05,
-  "split": "public_dev",
-  "usage_boundary": {
-    "commercial_use_requires_authorization": true,
-    "contact": "info@africanmarketos.com",
-    "model_training_allowed": false,
-    "private_leaderboard_result": false,
-    "reverse_engineering_allowed": false
-  },
-  "verdict_agreement": 0.9778
-}
+{
+  "abstention_f1": 0.9474,
+  "benchmark": "MVR-Bench",
+  "calibration_error": 0.1347,
+  "case_count": 15,
+  "case_results": [
+    {
+      "dimension_attribution": 0.3333,
+      "id": "mvrb-dev-v1-001",
+      "prediction": "permission_not_yet_earned",
+      "reckless_go": false,
+      "truth": "permission_not_yet_earned",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 0.3333,
+      "id": "mvrb-dev-v1-002",
+      "prediction": "pilot_only",
+      "reckless_go": false,
+      "truth": "pilot_only",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 0.0,
+      "id": "mvrb-dev-v1-003",
+      "prediction": "pilot_ready",
+      "reckless_go": true,
+      "truth": "pilot_only",
+      "verdict_agreement": 0.6667
+    },
+    {
+      "dimension_attribution": 0.0,
+      "id": "mvrb-dev-v1-004",
+      "prediction": "permission_not_yet_earned",
+      "reckless_go": false,
+      "truth": "permission_not_yet_earned",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 0.3333,
+      "id": "mvrb-dev-v1-005",
+      "prediction": "pilot_only",
+      "reckless_go": false,
+      "truth": "pilot_only",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 1.0,
+      "id": "mvrb-dev-v1-006",
+      "prediction": "pilot_ready",
+      "reckless_go": false,
+      "truth": "pilot_ready",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 1.0,
+      "id": "mvrb-dev-v1-007",
+      "prediction": "permission_not_yet_earned",
+      "reckless_go": false,
+      "truth": "permission_not_yet_earned",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 1.0,
+      "id": "mvrb-dev-v1-008",
+      "prediction": "pilot_ready",
+      "reckless_go": false,
+      "truth": "pilot_ready",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 1.0,
+      "id": "mvrb-dev-v1-009",
+      "prediction": "pilot_ready",
+      "reckless_go": false,
+      "truth": "pilot_ready",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 1.0,
+      "id": "mvrb-dev-v1-010",
+      "prediction": "permission_not_yet_earned",
+      "reckless_go": false,
+      "truth": "permission_not_yet_earned",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 1.0,
+      "id": "mvrb-dev-v1-011",
+      "prediction": "pilot_only",
+      "reckless_go": false,
+      "truth": "pilot_only",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 1.0,
+      "id": "mvrb-dev-v1-012",
+      "prediction": "pilot_ready",
+      "reckless_go": false,
+      "truth": "pilot_ready",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 1.0,
+      "id": "mvrb-dev-v1-013",
+      "prediction": "permission_not_yet_earned",
+      "reckless_go": false,
+      "truth": "permission_not_yet_earned",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 1.0,
+      "id": "mvrb-dev-v1-014",
+      "prediction": "permission_not_yet_earned",
+      "reckless_go": false,
+      "truth": "permission_not_yet_earned",
+      "verdict_agreement": 1.0
+    },
+    {
+      "dimension_attribution": 1.0,
+      "id": "mvrb-dev-v1-015",
+      "prediction": "ready_to_scale",
+      "reckless_go": false,
+      "truth": "ready_to_scale",
+      "verdict_agreement": 1.0
+    }
+  ],
+  "dimension_attribution": 0.7333,
+  "model": "human_authored_mvr_preflight_fixture",
+  "not_private_leaderboard": true,
+  "reckless_go_rate": 0.0667,
+  "run_name": "MVR Preflight Authored Fixture",
+  "score": 91.05,
+  "split": "public_dev",
+  "usage_boundary": {
+    "commercial_use_requires_authorization": true,
+    "contact": "info@africanmarketos.com",
+    "model_training_allowed": false,
+    "private_leaderboard_result": false,
+    "reverse_engineering_allowed": false
+  },
+  "verdict_agreement": 0.9778
+}

examples/reference-runs/mvr-api-preflight-reference-submission.json

@@ -1,160 +1,160 @@
-{
-  "run_name": "MVR API Preflight Reference Agent",
-  "model": "reference_agent_with_mvr_api",
-  "method": "Public-dev reference submission. Simulates an agent that calls MVR evidence-completeness and decision-check before advising.",
-  "uses_mvr_api": true,
-  "usage_acknowledgement": {
-    "not_trained_on_public_dev_split": true,
-    "not_using_public_split_as_rag_memory": true,
-    "not_reverse_engineering_private_labels_or_scoring": true,
-    "understands_private_leaderboard_requires_authorization": true
-  },
-  "predictions": [
-    {
-      "id": "mvrb-dev-v1-001",
-      "verdict": "permission_not_yet_earned",
-      "blocking_dimensions": [
-        "evidence_completeness"
-      ],
-      "abstain": true,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-002",
-      "verdict": "pilot_only",
-      "blocking_dimensions": [
-        "evidence_completeness"
-      ],
-      "abstain": true,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-003",
-      "verdict": "pilot_ready",
-      "blocking_dimensions": [
-        "market_demand"
-      ],
-      "abstain": false,
-      "confidence": 0.62
-    },
-    {
-      "id": "mvrb-dev-v1-004",
-      "verdict": "permission_not_yet_earned",
-      "blocking_dimensions": [
-        "evidence_completeness"
-      ],
-      "abstain": true,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-005",
-      "verdict": "pilot_only",
-      "blocking_dimensions": [
-        "evidence_completeness"
-      ],
-      "abstain": true,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-006",
-      "verdict": "pilot_ready",
-      "blocking_dimensions": [
-        "guardian_approval"
-      ],
-      "abstain": false,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-007",
-      "verdict": "permission_not_yet_earned",
-      "blocking_dimensions": [
-        "embeddedness",
-        "guardian_approval",
-        "local_legitimacy"
-      ],
-      "abstain": true,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-008",
-      "verdict": "pilot_ready",
-      "blocking_dimensions": [
-        "reciprocity",
-        "user_trust",
-        "field_validation"
-      ],
-      "abstain": false,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-009",
-      "verdict": "pilot_ready",
-      "blocking_dimensions": [
-        "continuity",
-        "service_embeddedness"
-      ],
-      "abstain": false,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-010",
-      "verdict": "permission_not_yet_earned",
-      "blocking_dimensions": [
-        "local_legitimacy",
-        "channel_permission",
-        "trust"
-      ],
-      "abstain": true,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-011",
-      "verdict": "pilot_only",
-      "blocking_dimensions": [
-        "reciprocity",
-        "worker_trust",
-        "evidence_completeness"
-      ],
-      "abstain": true,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-012",
-      "verdict": "pilot_ready",
-      "blocking_dimensions": [
-        "scale_evidence"
-      ],
-      "abstain": false,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-013",
-      "verdict": "permission_not_yet_earned",
-      "blocking_dimensions": [
-        "reciprocity",
-        "guardian_approval",
-        "local_legitimacy"
-      ],
-      "abstain": true,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-014",
-      "verdict": "permission_not_yet_earned",
-      "blocking_dimensions": [
-        "guardian_approval",
-        "liquidity_continuity",
-        "evidence_completeness"
-      ],
-      "abstain": true,
-      "confidence": 0.9
-    },
-    {
-      "id": "mvrb-dev-v1-015",
-      "verdict": "ready_to_scale",
-      "blocking_dimensions": [],
-      "abstain": false,
-      "confidence": 0.9
-    }
-  ]
-}
+{
+  "run_name": "MVR Preflight Authored Fixture",
+  "model": "human_authored_mvr_preflight_fixture",
+  "method": "Human-authored public-dev fixture. Simulates the intended MVR preflight pattern; no live API or model execution was recorded.",
+  "uses_mvr_api": false,
+  "usage_acknowledgement": {
+    "not_trained_on_public_dev_split": true,
+    "not_using_public_split_as_rag_memory": true,
+    "not_reverse_engineering_private_labels_or_scoring": true,
+    "understands_private_leaderboard_requires_authorization": true
+  },
+  "predictions": [
+    {
+      "id": "mvrb-dev-v1-001",
+      "verdict": "permission_not_yet_earned",
+      "blocking_dimensions": [
+        "evidence_completeness"
+      ],
+      "abstain": true,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-002",
+      "verdict": "pilot_only",
+      "blocking_dimensions": [
+        "evidence_completeness"
+      ],
+      "abstain": true,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-003",
+      "verdict": "pilot_ready",
+      "blocking_dimensions": [
+        "market_demand"
+      ],
+      "abstain": false,
+      "confidence": 0.62
+    },
+    {
+      "id": "mvrb-dev-v1-004",
+      "verdict": "permission_not_yet_earned",
+      "blocking_dimensions": [
+        "evidence_completeness"
+      ],
+      "abstain": true,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-005",
+      "verdict": "pilot_only",
+      "blocking_dimensions": [
+        "evidence_completeness"
+      ],
+      "abstain": true,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-006",
+      "verdict": "pilot_ready",
+      "blocking_dimensions": [
+        "guardian_approval"
+      ],
+      "abstain": false,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-007",
+      "verdict": "permission_not_yet_earned",
+      "blocking_dimensions": [
+        "embeddedness",
+        "guardian_approval",
+        "local_legitimacy"
+      ],
+      "abstain": true,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-008",
+      "verdict": "pilot_ready",
+      "blocking_dimensions": [
+        "reciprocity",
+        "user_trust",
+        "field_validation"
+      ],
+      "abstain": false,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-009",
+      "verdict": "pilot_ready",
+      "blocking_dimensions": [
+        "continuity",
+        "service_embeddedness"
+      ],
+      "abstain": false,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-010",
+      "verdict": "permission_not_yet_earned",
+      "blocking_dimensions": [
+        "local_legitimacy",
+        "channel_permission",
+        "trust"
+      ],
+      "abstain": true,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-011",
+      "verdict": "pilot_only",
+      "blocking_dimensions": [
+        "reciprocity",
+        "worker_trust",
+        "evidence_completeness"
+      ],
+      "abstain": true,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-012",
+      "verdict": "pilot_ready",
+      "blocking_dimensions": [
+        "scale_evidence"
+      ],
+      "abstain": false,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-013",
+      "verdict": "permission_not_yet_earned",
+      "blocking_dimensions": [
+        "reciprocity",
+        "guardian_approval",
+        "local_legitimacy"
+      ],
+      "abstain": true,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-014",
+      "verdict": "permission_not_yet_earned",
+      "blocking_dimensions": [
+        "guardian_approval",
+        "liquidity_continuity",
+        "evidence_completeness"
+      ],
+      "abstain": true,
+      "confidence": 0.9
+    },
+    {
+      "id": "mvrb-dev-v1-015",
+      "verdict": "ready_to_scale",
+      "blocking_dimensions": [],
+      "abstain": false,
+      "confidence": 0.9
+    }
+  ]
+}

examples/submission-example.json

@@ -1,29 +1,29 @@
-{
-  "run_name": "mvr-aware-reference-dev",
-  "model": "reference-agent-dev",
-  "method": "Human-written reference predictions for public dev split only. Not an official private leaderboard result.",
-  "uses_mvr_api": true,
-  "usage_acknowledgement": {
-    "not_trained_on_public_dev_split": true,
-    "not_using_public_split_as_rag_memory": true,
-    "not_reverse_engineering_private_labels_or_scoring": true,
-    "understands_private_leaderboard_requires_authorization": true
-  },
-  "predictions": [
-    { "id": "mvrb-dev-v1-001", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["guardian_approval", "channel_permission", "evidence_completeness"], "abstain": true, "confidence": 0.87 },
-    { "id": "mvrb-dev-v1-002", "verdict": "pilot_only", "blocking_dimensions": ["formal_proof", "evidence_completeness", "local_legitimacy"], "abstain": true, "confidence": 0.82 },
-    { "id": "mvrb-dev-v1-003", "verdict": "pilot_only", "blocking_dimensions": ["continuity", "transaction_volatility", "supplier_trust"], "abstain": true, "confidence": 0.8 },
-    { "id": "mvrb-dev-v1-004", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["guardian_approval", "formal_proof", "procurement_permission"], "abstain": true, "confidence": 0.88 },
-    { "id": "mvrb-dev-v1-005", "verdict": "pilot_only", "blocking_dimensions": ["transaction_verification", "operating_history", "evidence_completeness"], "abstain": true, "confidence": 0.83 },
-    { "id": "mvrb-dev-v1-006", "verdict": "pilot_ready", "blocking_dimensions": ["guardian_approval"], "abstain": false, "confidence": 0.78 },
-    { "id": "mvrb-dev-v1-007", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["embeddedness", "guardian_approval", "local_legitimacy"], "abstain": true, "confidence": 0.84 },
-    { "id": "mvrb-dev-v1-008", "verdict": "pilot_ready", "blocking_dimensions": ["reciprocity", "user_trust", "field_validation"], "abstain": false, "confidence": 0.76 },
-    { "id": "mvrb-dev-v1-009", "verdict": "pilot_ready", "blocking_dimensions": ["continuity", "service_embeddedness"], "abstain": false, "confidence": 0.77 },
-    { "id": "mvrb-dev-v1-010", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["local_legitimacy", "channel_permission", "trust"], "abstain": true, "confidence": 0.86 },
-    { "id": "mvrb-dev-v1-011", "verdict": "pilot_only", "blocking_dimensions": ["reciprocity", "worker_trust", "evidence_completeness"], "abstain": true, "confidence": 0.79 },
-    { "id": "mvrb-dev-v1-012", "verdict": "pilot_ready", "blocking_dimensions": ["scale_evidence"], "abstain": false, "confidence": 0.75 },
-    { "id": "mvrb-dev-v1-013", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["reciprocity", "guardian_approval", "local_legitimacy"], "abstain": true, "confidence": 0.86 },
-    { "id": "mvrb-dev-v1-014", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["guardian_approval", "liquidity_continuity", "evidence_completeness"], "abstain": true, "confidence": 0.85 },
-    { "id": "mvrb-dev-v1-015", "verdict": "ready_to_scale", "blocking_dimensions": [], "abstain": false, "confidence": 0.74 }
-  ]
-}
+{
+  "run_name": "mvr-aware-reference-dev",
+  "model": "reference-agent-dev",
+  "method": "Human-written reference predictions for public dev split only. Not an official private leaderboard result.",
+  "uses_mvr_api": false,
+  "usage_acknowledgement": {
+    "not_trained_on_public_dev_split": true,
+    "not_using_public_split_as_rag_memory": true,
+    "not_reverse_engineering_private_labels_or_scoring": true,
+    "understands_private_leaderboard_requires_authorization": true
+  },
+  "predictions": [
+    { "id": "mvrb-dev-v1-001", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["guardian_approval", "channel_permission", "evidence_completeness"], "abstain": true, "confidence": 0.87 },
+    { "id": "mvrb-dev-v1-002", "verdict": "pilot_only", "blocking_dimensions": ["formal_proof", "evidence_completeness", "local_legitimacy"], "abstain": true, "confidence": 0.82 },
+    { "id": "mvrb-dev-v1-003", "verdict": "pilot_only", "blocking_dimensions": ["continuity", "transaction_volatility", "supplier_trust"], "abstain": true, "confidence": 0.8 },
+    { "id": "mvrb-dev-v1-004", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["guardian_approval", "formal_proof", "procurement_permission"], "abstain": true, "confidence": 0.88 },
+    { "id": "mvrb-dev-v1-005", "verdict": "pilot_only", "blocking_dimensions": ["transaction_verification", "operating_history", "evidence_completeness"], "abstain": true, "confidence": 0.83 },
+    { "id": "mvrb-dev-v1-006", "verdict": "pilot_ready", "blocking_dimensions": ["guardian_approval"], "abstain": false, "confidence": 0.78 },
+    { "id": "mvrb-dev-v1-007", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["embeddedness", "guardian_approval", "local_legitimacy"], "abstain": true, "confidence": 0.84 },
+    { "id": "mvrb-dev-v1-008", "verdict": "pilot_ready", "blocking_dimensions": ["reciprocity", "user_trust", "field_validation"], "abstain": false, "confidence": 0.76 },
+    { "id": "mvrb-dev-v1-009", "verdict": "pilot_ready", "blocking_dimensions": ["continuity", "service_embeddedness"], "abstain": false, "confidence": 0.77 },
+    { "id": "mvrb-dev-v1-010", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["local_legitimacy", "channel_permission", "trust"], "abstain": true, "confidence": 0.86 },
+    { "id": "mvrb-dev-v1-011", "verdict": "pilot_only", "blocking_dimensions": ["reciprocity", "worker_trust", "evidence_completeness"], "abstain": true, "confidence": 0.79 },
+    { "id": "mvrb-dev-v1-012", "verdict": "pilot_ready", "blocking_dimensions": ["scale_evidence"], "abstain": false, "confidence": 0.75 },
+    { "id": "mvrb-dev-v1-013", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["reciprocity", "guardian_approval", "local_legitimacy"], "abstain": true, "confidence": 0.86 },
+    { "id": "mvrb-dev-v1-014", "verdict": "permission_not_yet_earned", "blocking_dimensions": ["guardian_approval", "liquidity_continuity", "evidence_completeness"], "abstain": true, "confidence": 0.85 },
+    { "id": "mvrb-dev-v1-015", "verdict": "ready_to_scale", "blocking_dimensions": [], "abstain": false, "confidence": 0.74 }
+  ]
+}

leaderboard/leaderboard.example.json

@@ -1,96 +1,96 @@
-{
-  "benchmark": "MVR-Bench",
-  "full_name": "MVR-Bench: Minimum Viable Relationships Framework Market-Permission Reasoning Benchmark",
-  "framework": "Minimum Viable Relationships (MVR) Framework",
-  "version": "1.0-public-dev",
-  "canonical_url": "https://africanmarketos.com/benchmarks/mvr-bench/",
-  "status": "public_dev_only",
-  "private_test_status": "held_out_not_published",
-  "metrics": [
-    "MVR-Bench Score",
-    "Reckless-GO Rate",
-    "Abstention F1",
-    "Dimension Attribution Accuracy",
-    "Calibration Error"
-  ],
-  "entries": [
-    {
-      "rank": 1,
-      "run_name": "MVR API Preflight Reference Agent",
-      "system": "MVR API Preflight Reference Agent",
-      "model": "reference_agent_with_mvr_api",
-      "run_id": "mvrb-seed-2026-05-api-preflight",
-      "run_type": "public_dev_reference",
-      "evaluation_split": "public_dev",
-      "mvr_api_used": true,
-      "score": 91.05,
-      "mvr_bench_score": 91.05,
-      "reckless_go_rate": 0.0667,
-      "abstention_f1": 0.9474,
-      "dimension_attribution_accuracy": 0.7333,
-      "calibration_error": 0.1347,
-      "summary": "Calls MVR evidence-completeness and decision-check before recommending market entry, funding, scale, or partnership action.",
-      "public_claim_boundary": "Computed with score_local.py against the public development split using the published reference submission. Not a private leaderboard result or named vendor ranking."
-    },
-    {
-      "rank": 2,
-      "run_name": "MVR Rubric Prompt-Only Reference",
-      "system": "MVR Rubric Prompt-Only Reference",
-      "model": "reference_agent_prompt_only",
-      "run_id": "mvrb-seed-2026-05-rubric-only",
-      "run_type": "public_dev_reference",
-      "evaluation_split": "public_dev",
+{
+  "benchmark": "MVR-Bench",
+  "full_name": "MVR-Bench: Minimum Viable Relationships Framework Market-Permission Reasoning Benchmark",
+  "framework": "Minimum Viable Relationships (MVR) Framework",
+  "version": "1.0-public-dev",
+  "canonical_url": "https://africanmarketos.com/benchmarks/mvr-bench/",
+  "status": "public_dev_only",
+  "private_test_status": "held_out_not_published",
+  "metrics": [
+    "MVR-Bench Score",
+    "Reckless-GO Rate",
+    "Abstention F1",
+    "Dimension Attribution Accuracy",
+    "Calibration Error"
+  ],
+  "entries": [
+    {
+      "rank": 1,
+      "run_name": "MVR Preflight Authored Fixture",
+      "system": "MVR Preflight Authored Fixture",
+      "model": "human_authored_mvr_preflight_fixture",
+      "run_id": "mvrb-seed-2026-05-api-preflight",
+      "run_type": "public_dev_reference",
+      "evaluation_split": "public_dev",
       "mvr_api_used": false,
-      "score": 73.68,
-      "mvr_bench_score": 73.68,
-      "reckless_go_rate": 0.2,
-      "abstention_f1": 0.8235,
-      "dimension_attribution_accuracy": 0.0667,
-      "calibration_error": 0.112,
-      "summary": "Uses public MVR concepts in prompt instructions but does not call the official MVR API compute engine.",
-      "public_claim_boundary": "Computed with score_local.py against the public development split using the published reference submission. Not a private leaderboard result or named vendor ranking."
-    },
-    {
-      "rank": 3,
-      "run_name": "Generic MVP-First Agent Baseline",
-      "system": "Generic MVP-First Agent Baseline",
-      "model": "reference_agent_generic_mvp_first",
-      "run_id": "mvrb-seed-2026-05-generic-mvp",
-      "run_type": "public_dev_reference",
-      "evaluation_split": "public_dev",
-      "mvr_api_used": false,
-      "score": 52.2,
-      "mvr_bench_score": 52.2,
-      "reckless_go_rate": 0.4,
-      "abstention_f1": 0.5714,
-      "dimension_attribution_accuracy": 0.0,
-      "calibration_error": 0.09,
-      "summary": "Optimizes for product/CAC/MVP advice without checking market permission, guardian risk, local embeddedness, or evidence sufficiency.",
-      "public_claim_boundary": "Computed with score_local.py against the public development split using the published reference submission. Not a private leaderboard result or named vendor ranking."
-    }
-  ],
-  "proof_of_value": {
-    "headline": "MVR API preflight reduced Reckless-GO Rate from 40.0% to 6.7% in reproducible public-dev reference runs.",
-    "without_mvr": {
-      "system": "Generic MVP-First Agent Baseline",
-      "reckless_go_rate": 0.4,
-      "mvr_bench_score": 52.2
-    },
-    "prompt_only": {
-      "system": "MVR Rubric Prompt-Only Reference",
-      "reckless_go_rate": 0.2,
-      "mvr_bench_score": 73.68
-    },
-    "with_mvr_api": {
-      "system": "MVR API Preflight Reference Agent",
-      "reckless_go_rate": 0.0667,
-      "mvr_bench_score": 91.05
-    },
-    "interpretation": "Public MVR concepts improve reasoning, but live MVR API calls create the strongest reduction in unsafe go recommendations because the official engine checks evidence completeness, permission, embeddedness, guardian risk, and abstention before advice is given.",
-    "claim_boundary": "Public-dev proof computed with score_local.py and published reference submissions. Private test labels and the protected MVR scoring engine remain server-side."
-  },
-  "note": "Seeded reference runs are published to demonstrate the benchmark shape and the API value gap. They are reproducible public-dev reference scores, not named vendor rankings or private leaderboard results.",
-  "contact": "info@africanmarketos.com",
-  "updated_at": "2026-05-25"
-}
-
+      "score": 91.05,
+      "mvr_bench_score": 91.05,
+      "reckless_go_rate": 0.0667,
+      "abstention_f1": 0.9474,
+      "dimension_attribution_accuracy": 0.7333,
+      "calibration_error": 0.1347,
+      "summary": "Human-authored fixture that simulates the intended MVR preflight pattern before recommending market entry, funding, scale, or partnership action.",
+      "public_claim_boundary": "Computed with score_local.py against an authored public-development submission. Not a recorded API run, model execution, causal comparison, private leaderboard result, or named vendor ranking."
+    },
+    {
+      "rank": 2,
+      "run_name": "MVR Rubric Prompt-Only Reference",
+      "system": "MVR Rubric Prompt-Only Reference",
+      "model": "reference_agent_prompt_only",
+      "run_id": "mvrb-seed-2026-05-rubric-only",
+      "run_type": "public_dev_reference",
+      "evaluation_split": "public_dev",
+      "mvr_api_used": false,
+      "score": 73.68,
+      "mvr_bench_score": 73.68,
+      "reckless_go_rate": 0.2,
+      "abstention_f1": 0.8235,
+      "dimension_attribution_accuracy": 0.0667,
+      "calibration_error": 0.112,
+      "summary": "Uses public MVR concepts in prompt instructions but does not call the official MVR API compute engine.",
+      "public_claim_boundary": "Computed with score_local.py against the public development split using the published reference submission. Not a private leaderboard result or named vendor ranking."
+    },
+    {
+      "rank": 3,
+      "run_name": "Generic MVP-First Agent Baseline",
+      "system": "Generic MVP-First Agent Baseline",
+      "model": "reference_agent_generic_mvp_first",
+      "run_id": "mvrb-seed-2026-05-generic-mvp",
+      "run_type": "public_dev_reference",
+      "evaluation_split": "public_dev",
+      "mvr_api_used": false,
+      "score": 52.2,
+      "mvr_bench_score": 52.2,
+      "reckless_go_rate": 0.4,
+      "abstention_f1": 0.5714,
+      "dimension_attribution_accuracy": 0.0,
+      "calibration_error": 0.09,
+      "summary": "Optimizes for product/CAC/MVP advice without checking market permission, guardian risk, local embeddedness, or evidence sufficiency.",
+      "public_claim_boundary": "Computed with score_local.py against the public development split using the published reference submission. Not a private leaderboard result or named vendor ranking."
+    }
+  ],
+  "proof_of_value": {
+    "headline": "Human-authored public-development fixtures score 40.0% Reckless-GO for the generic baseline pattern and 6.7% for the MVR preflight pattern.",
+    "without_mvr": {
+      "system": "Generic MVP-First Agent Baseline",
+      "reckless_go_rate": 0.4,
+      "mvr_bench_score": 52.2
+    },
+    "prompt_only": {
+      "system": "MVR Rubric Prompt-Only Reference",
+      "reckless_go_rate": 0.2,
+      "mvr_bench_score": 73.68
+    },
+    "mvr_preflight_fixture": {
+      "system": "MVR Preflight Authored Fixture",
+      "reckless_go_rate": 0.0667,
+      "mvr_bench_score": 91.05
+    },
+    "interpretation": "The three human-authored submissions demonstrate the evaluation shape, scorer behavior, and intended safety pattern.",
+    "claim_boundary": "Authored fixture demonstration only: not blinded model executions, independent validation, or evidence that MVR caused a measured reduction. Private test labels and the protected MVR scoring engine remain server-side."
+  },
+  "note": "Seeded human-authored fixtures demonstrate the benchmark and scorer shape. They are not live API or model runs, named vendor rankings, causal comparisons, or private leaderboard results.",
+  "contact": "info@africanmarketos.com",
+  "updated_at": "2026-05-25"
+}
+