nature-reviewer-skills v2.1.0

Merge pull request #7 from GeoGeekLab/docs/v2.1-skill-readmes

docs: update domain skill READMEs for v2.1

nature-reviewer-skills v2.0.1

Overview

Nature Reviewer Skills v2.0.1 is a major architecture, evaluation, reliability, and security release.

This version introduces a shared reviewer core, repository-level continuous integration, structured reviewer-quality evaluation, improved reviewer-pattern retrieval, hardened manuscript processing, and standardized package contracts across all supported scientific domains.

v2.0.1 is the canonical release containing the new v2 architecture.

Highlights

Shared reviewer core

• Added the src/nature_reviewer_core package.
• Consolidated manuscript extraction, pattern retrieval, validation, report rendering, and evaluation logic.
• Reduced duplicated implementations across individual reviewer-skill packages.
• Standardized shared interfaces, configuration, and package contracts.

Improved reviewer-pattern retrieval

• Replaced basic keyword-count matching with field-weighted BM25 ranking.
• Added phrase weighting and query expansion.
• Added retrieval-confidence indicators.
• Added diversity-aware result selection to reduce repetitive reviewer concerns.
• Improved field-level length normalization and ranking behavior.

Reviewer-quality evaluation

• Added structured benchmark schemas and evaluation utilities.
• Added major-issue recall measurement.
• Added precision, recall, and F1 metrics.
• Added severity-agreement evaluation.
• Added evidence-anchor coverage measurement.
• Added cross-reviewer redundancy analysis.
• Added per-domain evaluation summaries.

Continuous integration and engineering quality

• Added root-level GitHub Actions workflows.
• Added testing across supported Python versions.
• Added repository-wide package validation.
• Added Ruff linting and formatting checks.
• Added strict MyPy checks for the shared core.
• Added Bandit security scanning.
• Added independent validation for all seven reviewer skills.

Secure manuscript processing

• Added defensive DOCX XML parsing.
• Added PDF file-size and page-count limits.
• Added encrypted-PDF rejection.
• Added archive path-traversal protection.
• Added symbolic-link rejection.
• Added archive expansion-ratio limits.
• Added extracted-text length limits.
• Added atomic output writing.

Scientific reliability controls

• Strengthened claim–evidence alignment requirements.
• Added explicit prohibitions against fabricating figures, tables, page numbers, references, data, results, or misconduct allegations.
• Added differentiated responsibilities for multiple simulated reviewers.
• Added duplicate-concern detection across reviewer reports.
• Clarified that simulated reviewers are not statistically independent expert reviewers.
• Strengthened the distinction between correlation, causation, mechanism, and unsupported extrapolation.

Repository standardization

• Standardized package manifests and metadata.
• Aligned license declarations.
• Added versioned documentation and migration guidance.
• Added pinned development and documentation dependencies.
• Preserved and normalized domain reviewer-pattern resources.
• Moved shared Python implementation to a conventional src layout.

Included reviewer skills

This release includes reviewer skills for:

• Atmospheric science
• Chemistry
• Climate and ecology
• Engineering
• Hydrology
• Materials science
• Remote sensing

Compatibility and migration

This release introduces substantial internal architecture changes.

Users who imported duplicated Python scripts directly from individual skill directories should migrate to the shared nature_reviewer_core interfaces.

Domain-specific SKILL.md files, reviewer guidance, gates, templates, and reviewer-pattern resources remain available.

Users who only install or invoke the reviewer skills through the documented skill interface should require fewer migration changes than users who depend on package-internal Python modules.

Validation

The release has been validated with:

• Package validation for all seven reviewer skills
• Shared-core automated tests
• Independent skill-package tests
• Ruff lint and formatting checks
• Strict MyPy type checking
• Bandit security scanning
• DOCX extraction and report-generation tests
• PDF anchor-extraction tests
• CLI validation and retrieval tests
• Benchmark-pipeline tests
• Clean archive extraction and regression validation

Evaluation limitation

The included benchmark validates the operation of the evaluation pipeline and its metrics.

Synthetic benchmark results must not be interpreted as evidence that the system has achieved expert-reviewer performance.

Independent domain-expert evaluation, blinded manuscript testing, negative controls, inter-annotator agreement analysis, calibration studies, and prospective validation remain necessary.

The reviewer skills should be used as structured scientific quality-control assistants, not as replacements for qualified domain experts, editors, or formal peer review.

Recommended upgrade

Users of v1.x are encouraged to upgrade when they need:

• Shared and reusable reviewer infrastructure
• Repository-level CI and validation
• More robust reviewer-pattern retrieval
• Structured reviewer-quality metrics
• Stronger manuscript-ingestion security
• More consistent behavior across scientific domains

Review the migration documentation before upgrading integrations that depend directly on internal scripts from individual reviewer-skill directories.

nature-reviewer-skills v1.0.1

nature-reviewer-skills v1.0.1

Initial public release of GeoGeekLab nature-reviewer-skills, a research-grade suite of Nature-style reviewer skills for claim-dependent scientific peer review across Earth system science, chemistry, engineering, and materials science.

This release organizes seven domain-specific reviewer skills into a coherent skill suite for pre-submission manuscript review, evidence-chain stress testing, reviewer-perspective simulation, and research-quality improvement.

The suite is built around one central idea:

High-level scientific review is not a checklist.
It is a disciplined stress test of claims, evidence chains, uncertainty, novelty, mechanism, validation, and scope.

---

What is included

This release contains the following skill packages.

Earth system science umbrella

nature-earth-system-reviewer-skills is an umbrella layer for four independent Earth-system reviewer skills:

• nature-remote-sensing-reviewer
• nature-atmospheric-science-reviewer
• nature-hydrology-reviewer
• nature-climate-ecology-reviewer

The Earth-system umbrella is designed for manuscripts involving Earth observation, atmospheric systems, hydrological processes, climate-ecology interactions, carbon and nutrient cycles, biodiversity, hydroclimatic risk, environmental monitoring, and sustainability-facing environmental inference.

It is not a merged mega-skill. Each Earth-system skill remains scientifically independent, with its own SKILL.md, reviewer-memory patterns, domain gates, scripts, templates, and tests.

Other domain reviewer skills

This release also includes three independent scientific reviewer skills:

• nature-chemistry-reviewer
• nature-engineering-reviewer
• nature-materials-science-reviewer

These skills extend the same reviewer-reasoning framework to chemical science, engineering systems, and materials science while preserving domain-specific standards for evidence, validation, mechanism, reproducibility, and application claims.

---

Repository structure

nature-reviewer-skills/
├── nature-earth-system-reviewer-skills/
│   └── skills/
│       ├── nature-remote-sensing-reviewer-skill/
│       ├── nature-atmospheric-science-reviewer-skill/
│       ├── nature-hydrology-reviewer-skill/
│       └── nature-climate-ecology-reviewer-skill/
├── nature-chemistry-reviewer-skill/
├── nature-engineering-reviewer-skill/
└── nature-materials-science-reviewer-skill/

Each skill package follows a GitHub-ready structure:

nature-*-reviewer-skill/
├── SKILL.md
├── README.md
├── MANIFEST.json
├── reviewer_db/
├── references/
├── templates/
├── scripts/
├── tests/
├── docs/
├── examples/
├── LICENSE
├── LICENSE-MIT
└── LICENSE-APACHE

---

Scientific design

The suite is designed around reviewer reasoning distillation, not generic writing assistance.

Instead of redistributing raw reviewer reports, the skills encode abstracted reviewer logic as reusable scientific reasoning patterns:

claim type → evidence risk → stress-test gate → reviewer concern → revision direction

The core review logic emphasizes:

• claim-dependent review routing
• evidence-chain stress testing
• domain-specific reviewer gates
• validation sufficiency
• uncertainty propagation
• mechanism and causality
• reproducibility and controls
• benchmark fairness
• scope discipline
• generalization boundaries
• revision-oriented critique

The intended output is a Nature-style pre-review: rigorous, domain-aware, critical, and useful for manuscript revision.

---

Domain coverage

Remote sensing and Earth observation

For satellite remote sensing, Earth observation products, retrieval models, geospatial machine learning, environmental mapping, spatial validation, product uncertainty, and trend inference.

Key stress tests include product validity, spatial generalization, out-of-domain conditions, leakage, uncertainty propagation, and the distinction between observed signal, retrieved variable, proxy, and scientific interpretation.

Atmospheric science

For atmospheric observation, reanalysis, weather and climate dynamics, aerosol-cloud-radiation interactions, atmospheric chemistry, extreme events, circulation change, and AI weather or climate systems.

Key stress tests include physical consistency, model dependence, forcing and boundary conditions, attribution versus correlation, scale coupling, and dynamical interpretation.

Hydrology and water systems

For catchment hydrology, river discharge, groundwater, soil moisture, floods, droughts, water quality, water resources, hydroclimatic extremes, and water-security implications.

Key stress tests include hydrological variable identity, water balance closure, station-to-basin transferability, concentration versus load, calibration and validation, equifinality, and policy-relevant overclaiming.

Climate, ecology, and sustainability-facing environmental science

For climate impacts, ecological response, biodiversity, ecosystem stability, carbon and nutrient cycles, land-use change, conservation inference, and coupled human-natural systems.

Key stress tests include causal inference, ecological mechanism, spatial confounding, climate-driver separation, temporal lags, resilience claims, carbon-cycle evidence, and sustainability implication strength.

Chemistry

For synthesis, catalysis, analytical chemistry, chemical biology, mechanistic chemistry, computational chemistry, and AI-enabled molecular discovery.

Key stress tests include compound identity, purity, controls, reproducibility, reaction scope, selectivity, catalytic metrics, mechanism, and computational-experimental consistency.

Engineering

For engineering systems, platforms, devices, robotics, biomedical engineering, environmental engineering, algorithms embodied in physical systems, and translational engineering claims.

Key stress tests include requirement-design-validation coherence, prototype versus platform claims, benchmark fairness, operating envelope, failure modes, deployment constraints, and real-world utility.

Materials science

For materials design, synthesis, processing, structure-property relationships, characterization, stability, degradation, performance benchmarking, and application-facing materials claims.

Key stress tests include material identity, structural evidence, synthesis-processing-property linkage, durability, mechanism of performance, benchmark comparability, scalability, and application boundaries.

---

What this release is for

This release is intended for:

• researchers preparing high-level journal submissions
• graduate students learning how top-tier peer review reasons
• principal investigators conducting internal manuscript pre-review
• scientific teams building manuscript quality-control workflows
• AI-assisted research systems requiring domain-aware review logic
• developers building structured scientific-review agents
• researchers interested in reviewer reasoning distillation and scientific evaluation systems

The suite is especially useful when a manuscript makes strong claims about mechanism, generalization, novelty, prediction, attribution, application readiness, or sustainability relevance.

---

What this release is not

This repository is not:

• an official Nature Portfolio resource
• a replacement for real peer review
• a guarantee of journal acceptance
• a generic writing-polish prompt library
• a redistributed peer-review corpus
• a collection of raw reviewer reports
• a tool for identifying or imitating individual reviewers

It is an independent research and engineering project for strengthening manuscripts before formal peer review.

---

Provenance and copyright boundary

The reviewer-memory layer is based on abstracted, non-verbatim reasoning patterns.

This release does not redistribute:

• raw peer-review PDFs
• long reviewer comments
• identifiable reviewer language
• full article text
• private or non-public review material

Only generalized review logic, domain gates, metadata structures, templates, scripts, and validation utilities are included.

---

Validation

The package is designed for local validation through included scripts.

Umbrella-level validation:

python nature-earth-system-reviewer-skills/scripts/validate_umbrella.py nature-earth-system-reviewer-skills
python -m compileall -q nature-earth-system-reviewer-skills/scripts
python -m pytest -q nature-earth-system-reviewer-skills/tests

Individual skill validation:

python <skill-folder>/scripts/validate_package.py <skill-folder>
python -m compileall -q <skill-folder>/scripts
python -m pytest -q <skill-folder>/tests

---

Release status

This is the first consolidated suite release.

It establishes:

• the top-level nature-reviewer-skills repository
• the Earth-system umbrella structure
• seven independent Nature-style reviewer skills
• package-level validation scripts
• consistent skill metadata and naming
• GitHub-ready repository organization
• a research-oriented public README and release structure

---

License

This repository is released under the MIT License.

Individual skill packages may include package-level license files and dual-license metadata where applicable.

---

Disclaimer

This project is independently developed by GeoGeekLab.

It is not affiliated with, endorsed by, or approved by Nature Portfolio, Springer Nature, or any journal publisher.

The skills are intended to support scientific self-review, manuscript improvement, reviewer-perspective simulation, and research-quality control. They should not be treated as editorial decisions, official peer-review outcomes, or publication guarantees.