[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EJ2EVC9CVTNVBMVB8Y5MHX":3,"skill-same:01M7EJ2EVC9CVTNVBMVB8Y5MHX":63},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":31,"commitSha":32,"skillDirectory":33,"forksCount":14,"repositoryHtmlUrl":34,"repositoryHomepageUrl":34,"ownerHtmlUrl":35,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":36,"install":37,"resources":50},"01M7EJ2EVC9CVTNVBMVB8Y5MHX","compositional-data-analysis",null,">- Transform compositional data (mole\u002Fmass fractions, GC compositions, sulfur speciation) using log-ratio transforms (clr, alr, ilr) and sequential binary partition (SBP)-based balances so regressions, PCA, and correlations are valid on the simplex. Use when analysing GC composition trends, sulfur-s","Mr-Nobody-Anonymous\u002FAll-skills","Mr-Nobody-Anonymous","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F184253119?v=4","01M7AEXFJHKTYQW5R0ZCQQTWFV","anonymous","01M7C00M3GQR4QKP9GZA42GFAK",0,"MIT","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fchemical-engineering\u002Fcompositional-data-analysis\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: compositional-data-analysis\ndescription: \">- Transform compositional data (mole\u002Fmass fractions, GC compositions, sulfur speciation) using log-ratio transforms (clr, alr, ilr) and sequential binary partition (SBP)-based balances so regressions, PCA, and correlations are valid on the simplex. Use when analysing GC composition trends, sulfur-s\"\ncategory: chemical-engineering\nversion: 1.0.0\ndisable-model-invocation: false\nrisk: low\nsource: \"https:\u002F\u002Fgithub.com\u002Fjskherman\u002Fengg-skills\"\nsource_repository: \"jskherman\u002Fengg-skills\"\nsource_path: \"skills\u002Fcompositional-data-analysis\u002FSKILL.md\"\nlicense: \"MIT\"\nimported_at: \"2026-09-20\"\n---\n# Compositional Data Analysis (CoDA)\n\n## Overview\n\nCompositional data — mole\u002Fmass fractions, GC compositions, sulfur\nspeciation, particle size distributions — live on the simplex. Raw\nregression and correlation on a simplex are mathematically invalid:\nspurious correlations arise from the unit-sum constraint, and standard\nstatistical methods do not have well-defined geometry there.\n\nThis skill provides:\n\n- **Centered log-ratio (clr)**: `clr_i = ln(x_i \u002F g(x))` with `g(x)` the\n  geometric mean.\n- **Additive log-ratio (alr)**: removes one component as denominator.\n- **Isometric log-ratio (ilr)**: orthonormal coordinates from a sequential\n  binary partition (SBP), giving balances with physical interpretation.\n- **Heavy-end vs body balance helper** (the canonical example from the LPG\n  sulfur problem: `H = {C5, C6+}`, `B = {C3, C4}`).\n\nIt also implements multiplicative zero-replacement for components that\nare reported as zero (real zeros are incompatible with log-ratio\ntransforms; rounded zeros below LOD must be imputed first).\n\n## Prerequisites\n\n1. `uv` available.\n2. On first use, the script writes `LICENSE_NOTIFICATION.txt`.\n\n## When to Use\n\n- Building a regression where the predictor (or response) is a composition.\n- Computing PCA \u002F clustering on multivariate compositions.\n- Constructing an interpretable \"balance\" for plant data\n  (heavy-end-vs-body, paraffin-vs-olefin, acid-vs-organic-sulfur).\n- Working with the LPG sulfur problem's `z_H` heavy-end balance.\n\n## Don't use for\n\n- Data that just happens to include some fractions; only use CoDA when the\n  vector lies on the simplex (sums to a constant).\n- Univariate analyses of a single component fraction — those are special\n  cases that often still need closure-aware interpretation but rarely full\n  CoDA machinery.\n- A substitute for physical modelling: ilr coordinates are abstract; map\n  back to balances with interpretable names.\n\n## Utility Scripts\n\n- `uv run scripts\u002Fcoda.py clr --x 0.4,0.5,0.05,0.05 --output \u002Ftmp\u002Fclr.json`\n- `uv run scripts\u002Fcoda.py alr --x 0.4,0.5,0.05,0.05 --denominator-index 0 --output \u002Ftmp\u002Falr.json`\n- `uv run scripts\u002Fcoda.py ilr-default --x 0.4,0.5,0.05,0.05 --output \u002Ftmp\u002Filr.json`\n- `uv run scripts\u002Fcoda.py ilr-sbp --x 0.4,0.5,0.05,0.05 --sbp \"1,-1,0,0;0,0,1,-1;1,1,-1,-1\" --output \u002Ftmp\u002Filr_sbp.json`\n- `uv run scripts\u002Fcoda.py heavy-end --composition c3=0.45,c4=0.45,c5=0.07,c6plus=0.03 --heavy c5,c6plus --body c3,c4 --output \u002Ftmp\u002Fheavy.json`\n- `uv run scripts\u002Fcoda.py zero-replace --x 0.0,0.30,0.65,0.05 --delta 1e-6 --output \u002Ftmp\u002Frepl.json`\n\n## Procedure\n\n1. Confirm the vector is compositional (sums to a constant; non-negative).\n2. Replace zeros (multiplicative replacement; `delta` typically 1e-6 to\n   1e-4 in mol\u002Fmass fraction).\n3. Choose the transform:\n   - clr if you want a symmetric, single-coordinate-per-component view.\n   - alr if you need to write a regression with one explicit reference.\n   - ilr (with a fixed SBP) if you want orthonormal, interpretable balances.\n4. Run downstream regression \u002F PCA \u002F correlation on the transformed\n   coordinates.\n5. Back-transform reported coefficients to balance interpretations\n   (heavy-vs-body, etc.).\n\n## Pitfalls\n\n- Running OLS on raw mole fractions; the unit-sum constraint induces\n  spurious correlations.\n- Treating measured zeros as exact (they are usually below-LOD); failing\n  to impute zeros and then taking logs.\n- Building an SBP after seeing the results (\"garden of forking paths\");\n  always lock the SBP before fitting.\n- Interpreting ilr coefficients directly. They are dimensionless balances;\n  the physical interpretation requires going back through the SBP.\n- Using clr in regression: clr coordinates sum to zero, so regressing\n  all of them yields a rank-deficient system. Use ilr or alr for\n  regression.\n- Confusing the heavy-end balance with a simple ratio. The balance\n  `coef * ln(g_H \u002F g_B)` with `coef = sqrt(r s \u002F (r + s))` carries\n  important geometric meaning.\n- Forgetting to scale the alr\u002Filr basis when comparing across studies; the\n  same SBP is essential for cross-study comparisons.\n\n## Fallback Strategies\n\n- If you cannot decide on an SBP, use the default Helmert basis. It is\n  orthonormal but the balances are not physically named — you must label\n  them after the fact.\n- If a component is structurally zero (never measurable), drop it from the\n  composition before transforming. Document this.\n\n## Verification\n\n- Run the listed script with representative inputs and an `--output` file when a deterministic calculation is available.\n- Confirm the JSON result contains `ok: true`, expected units, and no unhandled warnings.\n- Check result magnitudes against the stated assumptions, references, and a hand calculation or known operating range before reporting them.\n\n## References\n\n- `references\u002Ftransforms_and_geometry.md` — short, formulas-only summary.\n- Pawlowsky-Glahn, Egozcue, Tolosana-Delgado, *Modeling and Analysis of\n  Compositional Data*, Wiley, 2015.\n- Egozcue & Pawlowsky-Glahn, \"Groups of Parts and Their Balances in\n  Compositional Data Analysis,\" Mathematical Geology, 2005.\n\n## Anti-Patterns\n\n- Calling something a \"balance\" without specifying the SBP.\n- Reporting compositional coefficients in linear-scale units.\n- Running PCA on raw mol% data.\n","# Compositional Data Analysis (CoDA)\n\n## Overview\n\nCompositional data — mole\u002Fmass fractions, GC compositions, sulfur\nspeciation, particle size distributions — live on the simplex. Raw\nregression and correlation on a simplex are mathematically invalid:\nspurious correlations arise from the unit-sum constraint, and standard\nstatistical methods do not have well-defined geometry there.\n\nThis skill provides:\n\n- **Centered log-ratio (clr)**: `clr_i = ln(x_i \u002F g(x))` with `g(x)` the\n  geometric mean.\n- **Additive log-ratio (alr)**: removes one component as denominator.\n- **Isometric log-ratio (ilr)**: orthonormal coordinates from a sequential\n  binary partition (SBP), giving balances with physical interpretation.\n- **Heavy-end vs body balance helper** (the canonical example from the LPG\n  sulfur problem: `H = {C5, C6+}`, `B = {C3, C4}`).\n\nIt also implements multiplicative zero-replacement for components that\nare reported as zero (real zeros are incompatible with log-ratio\ntransforms; rounded zeros below LOD must be imputed first).\n\n## Prerequisites\n\n1. `uv` available.\n2. On first use, the script writes `LICENSE_NOTIFICATION.txt`.\n\n## When to Use\n\n- Building a regression where the predictor (or response) is a composition.\n- Computing PCA \u002F clustering on multivariate compositions.\n- Constructing an interpretable \"balance\" for plant data\n  (heavy-end-vs-body, paraffin-vs-olefin, acid-vs-organic-sulfur).\n- Working with the LPG sulfur problem's `z_H` heavy-end balance.\n\n## Don't use for\n\n- Data that just happens to include some fractions; only use CoDA when the\n  vector lies on the simplex (sums to a constant).\n- Univariate analyses of a single component fraction — those are special\n  cases that often still need closure-aware interpretation but rarely full\n  CoDA machinery.\n- A substitute for physical modelling: ilr coordinates are abstract; map\n  back to balances with interpretable names.\n\n## Utility Scripts\n\n- `uv run scripts\u002Fcoda.py clr --x 0.4,0.5,0.05,0.05 --output \u002Ftmp\u002Fclr.json`\n- `uv run scripts\u002Fcoda.py alr --x 0.4,0.5,0.05,0.05 --denominator-index 0 --output \u002Ftmp\u002Falr.json`\n- `uv run scripts\u002Fcoda.py ilr-default --x 0.4,0.5,0.05,0.05 --output \u002Ftmp\u002Filr.json`\n- `uv run scripts\u002Fcoda.py ilr-sbp --x 0.4,0.5,0.05,0.05 --sbp \"1,-1,0,0;0,0,1,-1;1,1,-1,-1\" --output \u002Ftmp\u002Filr_sbp.json`\n- `uv run scripts\u002Fcoda.py heavy-end --composition c3=0.45,c4=0.45,c5=0.07,c6plus=0.03 --heavy c5,c6plus --body c3,c4 --output \u002Ftmp\u002Fheavy.json`\n- `uv run scripts\u002Fcoda.py zero-replace --x 0.0,0.30,0.65,0.05 --delta 1e-6 --output \u002Ftmp\u002Frepl.json`\n\n## Procedure\n\n1. Confirm the vector is compositional (sums to a constant; non-negative).\n2. Replace zeros (multiplicative replacement; `delta` typically 1e-6 to\n   1e-4 in mol\u002Fmass fraction).\n3. Choose the transform:\n   - clr if you want a symmetric, single-coordinate-per-component view.\n   - alr if you need to write a regression with one explicit reference.\n   - ilr (with a fixed SBP) if you want orthonormal, interpretable balances.\n4. Run downstream regression \u002F PCA \u002F correlation on the transformed\n   coordinates.\n5. Back-transform reported coefficients to balance interpretations\n   (heavy-vs-body, etc.).\n\n## Pitfalls\n\n- Running OLS on raw mole fractions; the unit-sum constraint induces\n  spurious correlations.\n- Treating measured zeros as exact (they are usually below-LOD); failing\n  to impute zeros and then taking logs.\n- Building an SBP after seeing the results (\"garden of forking paths\");\n  always lock the SBP before fitting.\n- Interpreting ilr coefficients directly. They are dimensionless balances;\n  the physical interpretation requires going back through the SBP.\n- Using clr in regression: clr coordinates sum to zero, so regressing\n  all of them yields a rank-deficient system. Use ilr or alr for\n  regression.\n- Confusing the heavy-end balance with a simple ratio. The balance\n  `coef * ln(g_H \u002F g_B)` with `coef = sqrt(r s \u002F (r + s))` carries\n  important geometric meaning.\n- Forgetting to scale the alr\u002Filr basis when comparing across studies; the\n  same SBP is essential for cross-study comparisons.\n\n## Fallback Strategies\n\n- If you cannot decide on an SBP, use the default Helmert basis. It is\n  orthonormal but the balances are not physically named — you must label\n  them after the fact.\n- If a component is structurally zero (never measurable), drop it from the\n  composition before transforming. Document this.\n\n## Verification\n\n- Run the listed script with representative inputs and an `--output` file when a deterministic calculation is available.\n- Confirm the JSON result contains `ok: true`, expected units, and no unhandled warnings.\n- Check result magnitudes against the stated assumptions, references, and a hand calculation or known operating range before reporting them.\n\n## References\n\n- `references\u002Ftransforms_and_geometry.md` — short, formulas-only summary.\n- Pawlowsky-Glahn, Egozcue, Tolosana-Delgado, *Modeling and Analysis of\n  Compositional Data*, Wiley, 2015.\n- Egozcue & Pawlowsky-Glahn, \"Groups of Parts and Their Balances in\n  Compositional Data Analysis,\" Mathematical Geology, 2005.\n\n## Anti-Patterns\n\n- Calling something a \"balance\" without specifying the SBP.\n- Reporting compositional coefficients in linear-scale units.\n- Running PCA on raw mol% data.\n",{"name":5,"risk":24,"source":25,"license":15,"version":26,"category":27,"description":7,"imported_at":28,"source_path":29,"source_repository":30,"disable-model-invocation":20},"low","https:\u002F\u002Fgithub.com\u002Fjskherman\u002Fengg-skills","1.0.0","chemical-engineering","2026-09-20","skills\u002Fcompositional-data-analysis\u002FSKILL.md","jskherman\u002Fengg-skills","main","2d1750717747097b602acf30bd09a7cbd4b4c6ea","awesome_skills\u002Fchemical-engineering\u002Fcompositional-data-analysis","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous","2026-11-09 11:45:36",{"source":8,"skillName":5,"commands":38,"prompt":48,"zipAvailable":49},[39,42,45],{"tool":40,"command":41},"npx","npx skills add Mr-Nobody-Anonymous\u002FAll-skills --skill compositional-data-analysis",{"tool":43,"command":44},"bunx","bunx skills add Mr-Nobody-Anonymous\u002FAll-skills --skill compositional-data-analysis",{"tool":46,"command":47},"pnpm","pnpm dlx skills add Mr-Nobody-Anonymous\u002FAll-skills --skill compositional-data-analysis","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fchemical-engineering\u002Fcompositional-data-analysis\u002FSKILL.md",true,[51,57],{"path":52,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Ftransforms_and_geometry.md","REFERENCE","参考文档",2005,"https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fchemical-engineering\u002Fcompositional-data-analysis\u002Freferences\u002Ftransforms_and_geometry.md",{"path":58,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":61,"sourceHtmlUrl":62},"scripts\u002Fcoda.py","SCRIPT","脚本",4499,"https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fchemical-engineering\u002Fcompositional-data-analysis\u002Fscripts\u002Fcoda.py",[64,72,79,86,93,100],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":71},"01M7FFFZPC2025JQQHPHDX154J","dei-assessment","Assess diversity, equity, and inclusion with metrics. TRIGGERS - Use when user needs help with dei-assessment related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fdei-assessment\u002FSKILL.md",[],[],"2026-11-16 10:05:34",{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":71},"01M7FFFWN8H4WA7A6Y398QKB6J","dei-action-plan","Create DEI action plans with goals and initiatives. TRIGGERS - Use when user needs help with dei-action-plan related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fdei-action-plan\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":71},"01M7FFFSHTM0QZRVZ2NRJTJ0PY","culture-fit-assessment","Design culture add assessment processes. TRIGGERS - Use when user needs help with culture-fit-assessment related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fculture-fit-assessment\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":71},"01M7FFFMA12RKEQA38KHJSXXPS","cowork-to-code-bridge","Use an already-installed, independently verified cowork-to-code bridge to run narrowly approved actions on the user's own macOS, Linux, or WSL2 machine through a local file queue.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fcowork-to-code-bridge\u002FSKILL.md",[],[],{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":71},"01M7FFFH1F6M8YRGW5V487GH1K","counter-offer-response","Draft counter-offer responses for retention. TRIGGERS - Use when user needs help with counter-offer-response related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fcounter-offer-response\u002FSKILL.md",[],[],{"skillNo":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":71},"01M7FFFDGEHV8Y7J9N40HZXW52","continuous-feedback-system","Design continuous feedback systems. TRIGGERS - Use when user needs help with continuous-feedback-system related tasks.","https:\u002F\u002Fgithub.com\u002FMr-Nobody-Anonymous\u002FAll-skills\u002Fblob\u002Fmain\u002Fawesome_skills\u002Fhr-people\u002Fcontinuous-feedback-system\u002FSKILL.md",[],[]]