Skip to content

GeoGeekLab/nature-reviewer-skills / nature-earth-system-reviewer-skills/skills/polar-earth-system-review-orchestrator

polar-earth-system-review-orchestrator

Verified

Cross-disciplinary polar Earth-system manuscript review orchestrator for Arctic, Antarctic and Southern Ocean research. Routes claims to existing domain reviewer skills and applies polar-specific evidence gates for sparse observations, cryosphere variables, product dependence, conservation budgets, coupled attribution, paleoclimate archives and regional research responsibility.

FileKindSize
LICENSEother1.1 KB
MANIFEST.jsonconfig413 B
README.mdmarkdown5.4 KB
SKILL.mdskill md6.7 KB
benchmarks/README.mdmarkdown304 B
benchmarks/cases/polar-firn-conversion.jsonconfig550 B
benchmarks/cases/polar-glacier-local-generalization.jsonconfig579 B
benchmarks/cases/polar-icefree-threshold.jsonconfig579 B
benchmarks/cases/polar-indigenous-governance.jsonconfig623 B
benchmarks/cases/polar-negative-attribution.jsonconfig333 B
benchmarks/cases/polar-negative-budget-closure.jsonconfig315 B
benchmarks/cases/polar-negative-independent-validation.jsonconfig331 B
benchmarks/cases/polar-negative-permit-reporting.jsonconfig326 B
benchmarks/cases/polar-permafrost-variable.jsonconfig590 B
benchmarks/cases/polar-proxy-age.jsonconfig579 B
benchmarks/cases/polar-random-cv.jsonconfig539 B
benchmarks/cases/polar-seaice-sensor-transition.jsonconfig569 B
benchmarks/summary.jsonconfig454 B
checksums.sha256other5.3 KB
corpus/README.mdmarkdown200 B
corpus/distillation_report.mdmarkdown2.2 KB
corpus/index.csvother18.5 KB
corpus/sources.jsonlother26.6 KB
docs/SECURITY.mdmarkdown317 B
examples/evaluation_prediction.jsonconfig410 B
gates/attribution_coupling_nonstationarity.mdmarkdown2.9 KB
gates/ice_sheet_glacier_dynamics.mdmarkdown2.8 KB
gates/mass_energy_water_budget.mdmarkdown2.9 KB
gates/model_resolution_assimilation.mdmarkdown2.8 KB
gates/paleoclimate_proxy_chronology.mdmarkdown2.7 KB
gates/permafrost_ecology_biogeochemistry.mdmarkdown2.9 KB
gates/polar_remote_sensing_products.mdmarkdown2.9 KB
gates/regional_scope_seasonality.mdmarkdown3.0 KB
gates/research_responsibility_governance.mdmarkdown3.0 KB
gates/sampling_representativeness.mdmarkdown2.8 KB
gates/sea_ice_ocean_atmosphere.mdmarkdown2.9 KB
gates/uncertainty_statistics_validation.mdmarkdown2.7 KB
gates/variable_identity_comparability.mdmarkdown2.9 KB
references/architecture.mdmarkdown354 B
references/claim_dependent_gate_router.mdmarkdown402 B
references/evaluation_contract.mdmarkdown419 B
references/evidence_boundary.mdmarkdown400 B
references/source_methodology.mdmarkdown524 B
requirements-dev.txtmarkdown27 B
reviewer_db/patterns.csvother74.4 KB
reviewer_db/patterns.jsonlother161.4 KB
reviewer_db/summary.jsonconfig819 B
reviewer_roles.jsonconfig2.1 KB
router.jsonconfig4.6 KB
scripts/extract_text_with_anchors.pyscript1.2 KB
scripts/render_review_docx.pyscript971 B
scripts/reviewer_db.pyscript1.5 KB
scripts/route_review.pyscript836 B
scripts/validate_package.pyscript928 B
templates/review_report.mdmarkdown1.2 KB
tests/test_package.pyscript2.4 KB

gates/uncertainty_statistics_validation.md

View on GitHub
# Uncertainty Statistics Validation

Dependence, propagation, sensitivity, multiplicity, rare events, leakage, baselines and external validity.

## Internal questions

- **Serial and spatial dependence ignored:** Significance tests treat polar time series, pixels, stations, or model cells as independent. Resolution: Use dependence-aware uncertainty, report effective sample size, and test robustness across correlation models.
- **Uncertainty is not propagated to headline result:** Input, calibration, retrieval, age-model, density, forcing, or parameter uncertainties are listed but not propagated to the final trend, budget, or projection. Resolution: Propagate all material uncertainty sources, including covariance, and report their contribution to the final interval.
- **Sensitivity tests do not span plausible choices:** Robustness is claimed from a narrow set of thresholds, windows, products, priors, or model parameters selected near the preferred analysis. Resolution: Predefine a plausible sensitivity space, include alternative products and methods, and show which conclusions survive.
- **Multiple comparisons are unaccounted:** Large spatial, temporal, taxonomic, or parameter searches report selected significant patterns without multiplicity or selection correction. Resolution: Control the false-discovery process or use hierarchical models, report the search space, and validate selected patterns independently.
- **Rare-event probability is overprecise:** The date or probability of an ice-free day, tipping event, extreme melt, or circulation threshold is estimated from too few effective samples. Resolution: Report interval and threshold sensitivity, effective sample size, and conditional probability wording rather than a precise deterministic date.
- **Cross-validation leaks space or time:** Random splits place neighboring pixels, repeated sites, adjacent years, or derived samples in both train and test sets. Resolution: Use blocked spatial and temporal validation, hold out entire sites or regions, and report out-of-domain performance.
- **Baseline comparison is weak or mismatched:** A new method or model is compared with an outdated, differently forced, differently resolved, or insufficiently tuned baseline. Resolution: Use strong matched baselines, harmonize inputs and evaluation domains, and report uncertainty in performance differences.
- **Confidence language exceeds validation domain:** The manuscript claims robustness, ubiquity, prediction, or early warning despite narrow data, one event, or an in-sample fit. Resolution: Separate in-domain precision from transfer uncertainty and align language with the actual validation design.

Use only when relevant to a central claim. Do not expose this gate name or pattern IDs in the final referee report.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 323kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 323kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 323kUpdated Claude CodeCodexVerifiedInstall