# Environment Life Review Forge > Adapts evidence synthesis workflows for environmental, ecological, biomedical, and life-science questions. Use for PECO/PICO frameworks, exposure-outcome reviews, ecological heterogeneity, dose-response evidence, risk-of-bias planning, environmental indicators, NDVI or vegetation-index models, partial least squares regression, PLS VIP audits, ecosystem-service relationships, ESR synergy/trade-off… - **Type:** Skill - **Install:** `agentstack add skill-vambrocop-evidenceforge-environment-life-review-forge` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [Vambrocop](https://agentstack.voostack.com/s/vambrocop) - **Installs:** 0 - **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills) - **Latest version:** 0.1.0 - **License:** MIT - **Upstream author:** [Vambrocop](https://github.com/Vambrocop) - **Source:** https://github.com/Vambrocop/EvidenceForge/tree/main/skills/environment-life-review-forge ## Install ```sh agentstack add skill-vambrocop-evidenceforge-environment-life-review-forge ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # Environment Life Review Forge Use this skill for environmental, ecological, biomedical, and life-science systematic reviews where exposure, organism/population, outcome, context, and study design need careful domain adaptation. ## Core Principle Domain structure matters. The same effect-size workflow may be misleading if exposure windows, species, tissues, endpoints, geography, or measurement platforms are not comparable. ## Intake Identify: - domain: environment, ecology, toxicology, epidemiology, life science, molecular biology, public health; - framework: PECO or PICO; - population or organism; - exposure or intervention; - comparator; - outcomes/endpoints; - study design; - spatial and temporal scale; - ecosystem-service set, pairwise relationship definition, synergy/trade-off coding, and threshold-management target, if ecosystem-service relationships are in scope; - pollutant exposure, crop outcome, food-security endpoint, counterfactual air-quality target, and crop-calorie translation, if air-quality food-security modeling is in scope; - biodiversity dimension, stability metric, climate-stress gradient, and moderation/interaction target, if biodiversity-stability evidence is in scope; - spatial unit and aggregation boundary, if geospatial prediction is in scope; - minimum mapping unit and detection threshold, if small-patch systems are in scope; - target map variable, spatial resolution, observation inventory, predictor stack, spatial autocorrelation plan, and uncertainty layer, if an environmental map product is in scope; - measurement method; - bidirectional pathways, if impacts and feedbacks are both in scope; - expected heterogeneity. Load: - `references/environmental-life-science.md` for domain heterogeneity. - `references/cee-alignment.md` for environmental evidence standards. - `references/pls-vip-environmental-indicators.md` for NDVI, vegetation, soil, climate, ecological indicator, PLS regression, and VIP interpretation audits. - `references/ecosystem-service-threshold-ml.md` for ecosystem-service relationship mapping, GWR-plus-ML workflows, nonlinear driver interpretation, threshold/optimal-interval identification, and spatial management translation. - `references/air-quality-food-security.md` for ozone, aerosol, SIF, crop yield, crop-calorie, counterfactual air-quality targets, and food-security co-benefit modeling. - `references/soil-biodiversity-aridity-stability.md` for soil biodiversity, aridity gradients, ecosystem stability, climate-stress moderation, and biodiversity-function buffering claims. - `references/ant-soil-carbon-meta.md` for soil-fauna meta-analysis, ecosystem-engineer effects on SOC stock and CO2 flux, trait-mediated moderators, and climate-context extraction. - `references/small-wetland-methane-scaling.md` for wetland methane, small water bodies, fine-resolution remote sensing, and scale-sensitive upscaling. - `references/cryosphere-ground-ice-mapping.md` for permafrost, near-surface ground ice, borehole observations, geospatial predictors, ensemble machine learning, spatial autocorrelation, prediction intervals, and public map-data audits. - `references/agroecosystem-nutrient-meta-analysis.md` for crop yield, soil organic carbon, fertilizer, amendment, and nutrient-management meta-analyses. - `references/agricultural-ml-yield-prediction.md` for crop-yield prediction studies integrating meteorological, breeding, genomic, remote-sensing, or field-trial data. - `references/agricultural-irrigation-optimization.md` for brackish-water irrigation, water-salt-yield-emission trade-offs, GAM nonlinear response modeling, NSGA-II optimization, and decision ranges such as ECw management windows. - `references/environmental-causal-ml.md` for environmental causal machine learning studies using DML, CATE, AutoML, SHAP/PDP-style interpretation, high-dimensional pollutant exposure data, socioeconomic covariates, ARGs, drinking-water safety, or One Health outcomes. - `references/food-system-bidirectional-nexus.md` for food-system reviews linking environmental pressures, feedbacks, trade, diets, crops, livestock, and aquatic foods. - `references/food-waste-geospatial-ml.md` for county, city, supply-chain, or market-level food-waste forecasting with geospatial analytics and machine learning. - `references/environmental-scenario-synthesis.md` when a review builds a literature-derived database, machine-learning/spatial model, or policy scenario simulation. - `references/land-use-optimization-tradeoffs.md` when a study uses multiobjective optimization, Pareto frontiers, land-use allocation, or food-water-carbon trade-off modeling. - `references/system-hub-policy-synthesis.md` when a paper uses one focal variable, such as nitrogen, carbon, water, phosphorus, air pollution, or biodiversity pressure, to connect multiple environmental, production, health, or policy outcomes under a boundary or scenario framework. ## Workflow 1. Build PECO/PICO. 2. Define exposure or intervention precisely. 3. Define outcome families and measurement units. 4. Specify eligible designs. 5. Identify heterogeneity sources. 6. Plan risk-of-bias or study quality appraisal. 7. Plan grey-literature and supplementary search if relevant. 8. Decide narrative, evidence map, or meta-analysis. 9. Build domain-specific extraction table. Use `templates/peco-framework.md`. Use `templates/pls-vip-environmental-audit.md` for PLS/VIP environmental indicator studies. Use `templates/ecosystem-service-threshold-audit.md` and `templates/ecosystem-service-threshold-schema.csv` for ecosystem-service relationship threshold-management studies. Use `templates/air-quality-food-security-audit.md` and `templates/air-quality-food-security-schema.csv` for air-pollution, crop-yield, SIF, and food-security co-benefit studies. Use `templates/biodiversity-stability-climate-stress-audit.md` and `templates/biodiversity-stability-climate-stress-schema.csv` for soil biodiversity, aridity, ecosystem-stability, and climate-stress moderation studies. Use `templates/soil-fauna-carbon-meta-audit.md`, `templates/soil-fauna-carbon-schema.csv`, and `templates/soil-fauna-carbon-method-stack-schema.csv` for ant, termite, earthworm, or other soil-fauna meta-analyses that separate SOC stock, CO2 flux, organic-matter stability outcomes, and method-stack choices such as multilevel meta-analysis, random forest, and PLS-PM/path modeling. Use `templates/wetland-methane-scale-audit.md` and `templates/wetland-methane-geospatial-schema.csv` for small-wetland methane and scale-sensitive upscaling studies. Use `templates/cryosphere-ground-ice-map-audit.md` and `templates/cryosphere-map-validation-schema.csv` for permafrost, near-surface ground ice, and other cryosphere map products. Use `templates/food-environment-bidirectional-audit.md` and `templates/food-environment-pressure-schema.csv` for food-system nexus reviews. Use `templates/food-waste-forecast-audit.md` and `templates/food-waste-geospatial-feature-schema.csv` for geospatial food-waste forecasting. Use `templates/dual-outcome-meta-audit.md` and `templates/nutrient-meta-extraction-schema.csv` for agroecosystem nutrient meta-analysis. Use `templates/nutrient-meta-reproducibility-ledger.csv` when a nutrient meta-analysis provides Zenodo/OSF/GitHub data and code. Use `templates/nutrient-meta-dataset-schema.csv` and `templates/nutrient-meta-r-workflow-blueprint.csv` when designing data tables and R scripts for nutrient meta-analysis. Use `templates/ml-yield-prediction-audit.md` and `templates/ml-yield-feature-schema.csv` for agricultural ML yield-prediction studies. Use `templates/environmental-causal-ml-audit.md` and `templates/environmental-causal-ml-feature-schema.csv` for environmental causal ML studies. Use `templates/irrigation-optimization-audit.md` and `templates/irrigation-optimization-schema.csv` for brackish-water irrigation and water-salt-yield-emission optimization studies. Use `templates/scenario-model-audit.md` and `templates/policy-scenario-matrix.csv` for scenario-model evidence synthesis. Use `templates/pareto-frontier-audit.md` and `templates/multi-objective-tradeoff-schema.csv` for multiobjective optimization and land-use trade-off studies. Use `templates/system-hub-variable-audit.md` and `templates/system-hub-variable-schema.csv` for papers that organize evidence around a system hub variable, safe boundary, hotspot layer, co-benefit structure, or policy-portfolio translation. ## Output Modes ### PECO Protocol Memo ```text Population: Exposure: Comparator: Outcome families: Eligible designs: Heterogeneity: Risk of bias: Synthesis plan: ``` ### Domain Extraction Plan Include: - species/population; - site/geography; - exposure dose/intensity; - duration/window; - endpoint; - assay/measurement method; - confounders; - study quality. ### PLS/VIP Environmental Indicator Audit Use this mode when a study links NDVI, vegetation productivity, soil quality, biodiversity, ecosystem-service, pollutant, climate, or hydrological indicators to multiple correlated predictors using partial least squares regression and VIP rankings. Include: - outcome indicator and unit; - predictor families and expected ecological meaning; - sample size, time span, site count, and clustering; - missing-data and scaling decisions; - component-count selection rule; - cross-validation design and whether it respects time, site, or spatial grouping; - RMSEP, R2/Q2, residual checks, and influential observations; - VIP table, VIP threshold, and coefficient direction; - whether VIP is interpreted as predictive importance, mechanism, or causal effect; - robustness checks such as alternative component count, leave-one-year/site-out validation, or baseline regression comparison. ### Ecosystem-Service Threshold ML Audit Use this mode when a study maps ecosystem services or ecosystem-service relationships, then uses interpretable machine learning, GWR, GAM, XGBoost, GBDT, SHAP, PDP, ICE, ALE, or response-curve overlays to identify management thresholds or optimal driver intervals. Include: - ecosystem services and pairwise ESR definitions; - spatial unit, resolution, years, and service-assessment models; - relationship coding: synergy, trade-off, co-benefit, competition, or probability of comprehensive synergy; - driver families: climate, topography, landscape, land use, human pressure, accessibility, and policy; - spatial heterogeneity model such as GWR and its bandwidth/kernel choices; - ML model set, tuning, validation, and spatial leakage checks; - interpretability method and whether thresholds come from PDP, SHAP dependence, ALE, response-curve superposition, or another method; - optimal interval definition, probability target, and uncertainty; - zoning or management translation and whether it is descriptive, predictive, or causal; - robustness checks against alternative service models, spatial resolution, model family, and threshold rule. ### Air Quality Food-Security Audit Use this mode when a study estimates how ozone, PM2.5, aerosol optical depth, diffuse radiation, SIF, temperature, or related atmospheric conditions affect crop productivity, crop yield, calories, self-sufficiency, or food-security indicators. Include: - crops, regions, years, and spatial unit; - pollutant exposure metrics, such as AOT40, peak-season ozone, PM2.5, AOD, or aerosol loading; - crop outcome: yield, SIF, productivity proxy, calorie output, or supply-demand balance; - statistical model, flexible functional form, crop fixed effects, spatial effects, weather controls, and trend controls; - counterfactual air-quality targets and whether targets are policy-based; - nonlinear and synergistic pollutant-response claims; - validation against observed yields or independent crop productivity data; - translation from yield to calories, self-sufficiency, imports, or food security; - data/code availability and uncertainty in source data, crop area, crop-calorie conversion, and counterfactual scenarios. ### Biodiversity Stability Climate-Stress Audit Use this mode when a study evaluates whether biodiversity, soil biodiversity, microbial diversity, functional diversity, or community composition supports ecosystem stability under aridity, drought, warming, land-use stress, or other climate gradients. Include: - biodiversity dimension and measurement platform; - stability endpoint: temporal stability, resistance, resilience, multifunctionality, productivity variability, or service stability; - climate-stress gradient and exposure window; - ecosystem type, spatial domain, sampling design, and temporal depth; - interaction or moderation model: whether aridity weakens, strengthens, or changes the biodiversity-stability relationship; - controls for productivity, soil, climate, land use, management, and spatial dependence; - mechanism evidence from microbes, soil nutrients, plant traits, or food-web structure; - uncertainty, threshold, nonlinear, and subgroup evidence; - whether claims are observational associations, experiments, manipulations, or causal estimates. ### Soil Fauna Carbon Meta-Analysis Audit Use this mode when a study synthesizes how ants or other soil fauna affect soil carbon storage, carbon fluxes, organic matter turnover, or stability across ecosystems. Include: - focal fauna and functional traits; - stock versus flux versus stability outcome separation; - effect-size family and percent-change interpretation; - climate, latitude, baseline SOC, or ecosystem moderators; - dependence plan for multiple endpoints per study; - whether driver ranking uses random forest or another ML method; - whether indirect pathways are organized with PLS-PM, SEM, or another path model; - whether "more storage" and "more emissions" are both present and how the paper interprets that pattern. ### Small-Wetland Methane Scaling Audit Use this mode when a study estimates emissions from small wetlands, ponds, small water bodies, or patchy ecosystems where spatial resolution and minimum mapping unit change the global or regional budget. Include: - wetland or water-body size class; - mapping resolution and minimum detectable area; - forested/non-forested domain boundary; - wetland inventory or remote-sensing product; - flux model or emission-factor source; - annual trend window; - contribution to total emissions; - uncertainty, double-counting, and omission risks; - implications for methane budgets and restoration policy. ### Cryosphere Ground-Ice Map Audit Use this mode when a study creates or reuses a spatial map product for permafrost, near-surface ground ice, volumetric ice content, active-layer properties, thermokarst susceptibility, or related cryosphere hazards. Include: - target map variable and depth convention; - spatial domain, permafrost mask, map year/window, resolution, and grid definition; - field observation type, count, spatial distribution, and measurement comparability; - predictor stack: substrate, hydrology, topography, geology, paleoclimate, modern climate, remote sensing, and vegetation; - model families, ensemble strategy, calibration data, and simulation count; - validation design, including independent or spatially blocked validation; - spatial autocorrelation, sampling bias, and extrapolation checks; - accuracy, bias, RMSE, R-squared, prediction interval, and uncertainty maps; - storage, extent, hazard, infrastructure, climate, hydrology, and ecosystem interpretations; - data DOI, code availability, and versioning of map products. ### Agroecosystem Nutrient Meta-Analysis Use this mode for fertilizer, manure, compost, liming, biochar, and nutrient-management reviews with crop, soil, emission, or microbial outcomes. Include: - intervention nutrient form and rate; - comparator nutrient background; - crop or ecosystem; - soil baseline status; - climate and geography; - experiment duration; - yield endpoint; - soil-carbon or soil-health endpoint; - response-ratio or percent-change metric; - moderator plan; - … ## Source & license This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [Vambrocop](https://github.com/Vambrocop) - **Source:** [Vambrocop/EvidenceForge](https://github.com/Vambrocop/EvidenceForge) - **License:** MIT Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** no - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/skill-vambrocop-evidenceforge-environment-life-review-forge - Seller: https://agentstack.voostack.com/s/vambrocop - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.