[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7F5WN18S9KF15W14FJ2G732":3,"skill-same:01M7F5WN18S9KF15W14FJ2G732":65},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":29,"commitSha":30,"skillDirectory":31,"forksCount":13,"repositoryHtmlUrl":32,"repositoryHomepageUrl":6,"ownerHtmlUrl":33,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":34,"install":35,"resources":48},"01M7F5WN18S9KF15W14FJ2G732","mageck",null,"Analyzes pooled CRISPR screen FASTQ reads and guide-count matrices with MAGeCK, validates guide libraries and contrasts, measures replicate and library QC, and produces gene hit rankings with effect sizes and FDR. Use for new knockout, CRISPRi, or CRISPRa screen analysis, enrichment or depletion contrasts, and MAGeCK count\u002Ftest workflows; existing public dependency-score lookup belongs to DepMap.","beanspowt\u002Fbeancode","beanspowt","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F333943283?v=4","01M7F39F0GDWKS57BCKW8JJH7F","01M7F39F1B1TVPDJMS37BZQHWM",0,"MIT","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-science\u002Fskills\u002Fmageck\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: mageck\ndescription: Analyzes pooled CRISPR screen FASTQ reads and guide-count matrices with MAGeCK, validates guide libraries and contrasts, measures replicate and library QC, and produces gene hit rankings with effect sizes and FDR. Use for new knockout, CRISPRi, or CRISPRa screen analysis, enrichment or depletion contrasts, and MAGeCK count\u002Ftest workflows; existing public dependency-score lookup belongs to DepMap.\nlicense: MIT\ncompatibility: Requires Python 3.10+ for the helper and MAGeCK 0.5.9.5 with its compiled RRA executable for analysis. Tested runtime uses Python 3.11, NumPy 1.26.4 and SciPy 1.13.1. Source installation needs a C++ compiler; network is needed only for installation. No credentials.\nmetadata:\n  version: \"1.1\"\n  skill-author: K-Dense Inc.\n  upstream-version: \"0.5.9.5\"\n  last-reviewed: \"2026-10-01\"\n---\n\n# MAGeCK pooled-screen analysis\n\n## When to use\n\nUse this skill to count existing sequencing reads against a supplied guide library or compare\nalready-counted pooled screens. Deliver the count matrix, QC, guide and gene results, contrast\nprovenance, and a short interpretation of enrichment\u002Fdepletion. This workflow analyzes screens;\nit does not design guides or infer gene function from a hit alone.\n\n## Runtime\n\nThe tested source installation and external-runtime caveat are in\n[references\u002Fruntime.md](references\u002Fruntime.md). Verify both `mageck --version` and\n`mageck test --help` before an analysis. The bundled Python helper is standard-library only.\nMAGeCK itself also needs NumPy, SciPy, and the `RRA` binary. PDF\u002FR reporting is optional and not\nneeded by the helper.\n\nThe [official release directory](https:\u002F\u002Fsourceforge.net\u002Fprojects\u002Fmageck\u002Ffiles\u002F0.5\u002F) still lists\n0.5.9.5 as its latest MAGeCK release. Upstream now links the separate\n[MAGeCK2 project](https:\u002F\u002Fgithub.com\u002Fdavidliwei\u002Fmageck2); these commands and the helper target\nMAGeCK 0.5.9.5, not an interchangeable MAGeCK2 installation. This is a local CLI workflow with\nno service API or authentication.\n\n## Workflow\n\n1. Establish the library version, perturbation modality, sample names, selection direction,\n   biological replicates, baseline material, time point, and batch. Separate sequencing lanes\n   from independent biological replicates. A plasmid baseline and a cell day-zero baseline\n   answer different questions. Require an explicit treatment\u002Fcontrol contrast; the helper\n   never silently assigns all unused samples to the control group.\n2. Validate a **headerless TSV library** containing guide ID, DNA sequence, gene. IDs and\n   sequences must be unique. The helper requires a count-table header beginning `sgRNA`, `Gene`,\n   followed by unique nonnumeric sample names such as `c1`. MAGeCK can interpret numeric names\n   as column indices or count values; rename them before analysis. Guide\u002Fgene IDs must have no\n   whitespace. The helper rejects ambiguous sequences and any count\u002Flibrary ID or gene mismatch;\n   resolve intentional multi-target guides explicitly upstream. These are deliberate helper\n   restrictions; native MAGeCK also accepts other input variants.\n3. For FASTQ, inspect read structure and known guide sequences to establish trimming and\n   orientation. Use MAGeCK `count`, with one space-separated argument per biological sample;\n   comma-join lanes only when they are technical replicates of that same sample. Preserve\n   unmapped-read and count-summary evidence when mapping is poor. A zero-count guide remains\n   in the library; do not drop it to improve QC.\n4. Run QC before statistical testing. Review library representation, median reads per guide,\n   zero fractions, Gini coefficients, and within-condition replicate correlations. The\n   helper's 10% zero and 0.8 correlation flags are review prompts, not universal acceptance\n   thresholds. High correlation can coexist with systematic artifacts. Read depth is not\n   experimental cell coverage. The helper uses a raw-count population Gini; MAGeCK's native\n   count-summary Gini uses log(count + 1) with a finite-sample correction. Do not compare their\n   values or thresholds as the same statistic.\n5. Choose normalization based on the screen. Median normalization assumes most guides are\n   stable. For a strong global shift, supplied validated negative-control guides may support\n   `--normalization control`. These must be **guide IDs**, one per line; a gene list is not\n   interchangeable. Biological control samples and negative-control guides serve different\n   roles. At least two controls must be present, and every guide assigned to a control gene\n   must be designated a control. Supplying `--control-guides` also changes the RRA null distribution,\n   even with median normalization. Record their origin and check their count distribution.\n   MAGeCK 0.5.9.5 switches median normalization to total-count scaling for a zero median or\n   more than 45% zero guides in any selected sample; for control normalization it evaluates\n   the control-guide subset. Check the report's applied method, size factors, and warnings.\n6. Use `test` for a two-group comparison. `--paired` requires both lists in corresponding\n   biological order and equal length; matching lengths alone do not establish pairing.\n   The helper reports genes at the requested FDR in both directions and retains full rankings.\n   For a multi-factor design, see [references\u002Fdesign.md](references\u002Fdesign.md); do not collapse\n   batches or time courses into an unjustified two-group test.\n7. Inspect guide concordance for leading genes, essential-gene recovery where appropriate,\n   negative controls, replicate consistency, and copy-number artifacts in nuclease knockout\n   screens. Report effect sizes alongside FDR. An enriched guide can indicate resistance,\n   growth advantage, or a sampling artifact depending on the selection; depletion need not\n   imply universal essentiality. Lack of replication or low-count guides weakens inference.\n\n## Commands\n\nRun paths relative to the installed skill directory. Input and output paths refer to the user's\nanalysis directory. The helper refuses to reuse an existing results directory.\n\n```bash\n# Single-end guide reads; trim and orientation must match the user's library preparation.\nmageck count -l library.tsv --fastq c1.fastq.gz c2.fastq.gz t1.fastq.gz t2.fastq.gz \\\n  --sample-label c1,c2,t1,t2 --trim-5 0 --norm-method none -n counts\n\npython scripts\u002Fscreen_analysis.py qc --counts counts.count.txt --library library.tsv \\\n  --control c1 c2 --treatment t1 t2 --output qc.json\n\npython scripts\u002Fscreen_analysis.py test --counts counts.count.txt --library library.tsv \\\n  --control c1 c2 --treatment t1 t2 --normalization median --fdr 0.05 --output result\n```\n\nThe FASTQ command structure was exercised with a synthetic two-guide library: counts of 30 and\n12 were recovered exactly. A separate trimmed, reverse-complemented, two-lane fixture recovered\n15, 7, and 0 reads. The two-group helper was exercised with 500 guides and two replicates per\ncondition in unpaired and paired modes; known depleted and enriched genes ranked first in their\nrespective directions and passed FDR 0.05. Sparse fixtures verified the total-normalization fallback.\nThese tests establish execution and signal direction, not real-screen statistical calibration.\nReal-file paths above are illustrative.\n\n`result\u002Freport.json` records count\u002Flibrary\u002Fcontrol-guide SHA-256, control-guide IDs, MAGeCK version,\nactual arguments, QC, applied normalization, warnings, hit direction, FDR, log2 fold change and rank.\nThe helper explicitly selects median guide LFC aggregation and `--remove-zero both`: all-zero\nguides remain in the input\u002FQC but are excluded from ranking. Native MAGeCK also skips `NA`\u002F`na`\ngene labels by default. The helper's `--fdr` filters completed gene results; it does not set MAGeCK's\n`--gene-test-fdr-threshold`, which controls the RRA guide-selection cutoff. Negative and positive\nFDRs are separate families; their union does not establish joint FDR control across directions or\nacross multiple contrasts. `screen.gene_summary.txt`, `screen.sgrna_summary.txt`, normalized\ncounts and the execution log retain the complete evidence. Include the original library, sample\nsheet and negative-control list in the analysis handoff; the count checksum cannot reconstruct them.\n\n## Primary references\n\n- [MAGeCK usage and command semantics](https:\u002F\u002Fsourceforge.net\u002Fp\u002Fmageck\u002Fwiki\u002Fusage\u002F)\n- [Official installation guidance](https:\u002F\u002Fsourceforge.net\u002Fp\u002Fmageck\u002Fwiki\u002Finstall\u002F)\n- [Input formats](https:\u002F\u002Fsourceforge.net\u002Fp\u002Fmageck\u002Fwiki\u002Finput\u002F)\n- [Output field meanings](https:\u002F\u002Fsourceforge.net\u002Fp\u002Fmageck\u002Fwiki\u002Foutput\u002F)\n","# MAGeCK pooled-screen analysis\n\n## When to use\n\nUse this skill to count existing sequencing reads against a supplied guide library or compare\nalready-counted pooled screens. Deliver the count matrix, QC, guide and gene results, contrast\nprovenance, and a short interpretation of enrichment\u002Fdepletion. This workflow analyzes screens;\nit does not design guides or infer gene function from a hit alone.\n\n## Runtime\n\nThe tested source installation and external-runtime caveat are in\n[references\u002Fruntime.md](references\u002Fruntime.md). Verify both `mageck --version` and\n`mageck test --help` before an analysis. The bundled Python helper is standard-library only.\nMAGeCK itself also needs NumPy, SciPy, and the `RRA` binary. PDF\u002FR reporting is optional and not\nneeded by the helper.\n\nThe [official release directory](https:\u002F\u002Fsourceforge.net\u002Fprojects\u002Fmageck\u002Ffiles\u002F0.5\u002F) still lists\n0.5.9.5 as its latest MAGeCK release. Upstream now links the separate\n[MAGeCK2 project](https:\u002F\u002Fgithub.com\u002Fdavidliwei\u002Fmageck2); these commands and the helper target\nMAGeCK 0.5.9.5, not an interchangeable MAGeCK2 installation. This is a local CLI workflow with\nno service API or authentication.\n\n## Workflow\n\n1. Establish the library version, perturbation modality, sample names, selection direction,\n   biological replicates, baseline material, time point, and batch. Separate sequencing lanes\n   from independent biological replicates. A plasmid baseline and a cell day-zero baseline\n   answer different questions. Require an explicit treatment\u002Fcontrol contrast; the helper\n   never silently assigns all unused samples to the control group.\n2. Validate a **headerless TSV library** containing guide ID, DNA sequence, gene. IDs and\n   sequences must be unique. The helper requires a count-table header beginning `sgRNA`, `Gene`,\n   followed by unique nonnumeric sample names such as `c1`. MAGeCK can interpret numeric names\n   as column indices or count values; rename them before analysis. Guide\u002Fgene IDs must have no\n   whitespace. The helper rejects ambiguous sequences and any count\u002Flibrary ID or gene mismatch;\n   resolve intentional multi-target guides explicitly upstream. These are deliberate helper\n   restrictions; native MAGeCK also accepts other input variants.\n3. For FASTQ, inspect read structure and known guide sequences to establish trimming and\n   orientation. Use MAGeCK `count`, with one space-separated argument per biological sample;\n   comma-join lanes only when they are technical replicates of that same sample. Preserve\n   unmapped-read and count-summary evidence when mapping is poor. A zero-count guide remains\n   in the library; do not drop it to improve QC.\n4. Run QC before statistical testing. Review library representation, median reads per guide,\n   zero fractions, Gini coefficients, and within-condition replicate correlations. The\n   helper's 10% zero and 0.8 correlation flags are review prompts, not universal acceptance\n   thresholds. High correlation can coexist with systematic artifacts. Read depth is not\n   experimental cell coverage. The helper uses a raw-count population Gini; MAGeCK's native\n   count-summary Gini uses log(count + 1) with a finite-sample correction. Do not compare their\n   values or thresholds as the same statistic.\n5. Choose normalization based on the screen. Median normalization assumes most guides are\n   stable. For a strong global shift, supplied validated negative-control guides may support\n   `--normalization control`. These must be **guide IDs**, one per line; a gene list is not\n   interchangeable. Biological control samples and negative-control guides serve different\n   roles. At least two controls must be present, and every guide assigned to a control gene\n   must be designated a control. Supplying `--control-guides` also changes the RRA null distribution,\n   even with median normalization. Record their origin and check their count distribution.\n   MAGeCK 0.5.9.5 switches median normalization to total-count scaling for a zero median or\n   more than 45% zero guides in any selected sample; for control normalization it evaluates\n   the control-guide subset. Check the report's applied method, size factors, and warnings.\n6. Use `test` for a two-group comparison. `--paired` requires both lists in corresponding\n   biological order and equal length; matching lengths alone do not establish pairing.\n   The helper reports genes at the requested FDR in both directions and retains full rankings.\n   For a multi-factor design, see [references\u002Fdesign.md](references\u002Fdesign.md); do not collapse\n   batches or time courses into an unjustified two-group test.\n7. Inspect guide concordance for leading genes, essential-gene recovery where appropriate,\n   negative controls, replicate consistency, and copy-number artifacts in nuclease knockout\n   screens. Report effect sizes alongside FDR. An enriched guide can indicate resistance,\n   growth advantage, or a sampling artifact depending on the selection; depletion need not\n   imply universal essentiality. Lack of replication or low-count guides weakens inference.\n\n## Commands\n\nRun paths relative to the installed skill directory. Input and output paths refer to the user's\nanalysis directory. The helper refuses to reuse an existing results directory.\n\n```bash\n# Single-end guide reads; trim and orientation must match the user's library preparation.\nmageck count -l library.tsv --fastq c1.fastq.gz c2.fastq.gz t1.fastq.gz t2.fastq.gz \\\n  --sample-label c1,c2,t1,t2 --trim-5 0 --norm-method none -n counts\n\npython scripts\u002Fscreen_analysis.py qc --counts counts.count.txt --library library.tsv \\\n  --control c1 c2 --treatment t1 t2 --output qc.json\n\npython scripts\u002Fscreen_analysis.py test --counts counts.count.txt --library library.tsv \\\n  --control c1 c2 --treatment t1 t2 --normalization median --fdr 0.05 --output result\n```\n\nThe FASTQ command structure was exercised with a synthetic two-guide library: counts of 30 and\n12 were recovered exactly. A separate trimmed, reverse-complemented, two-lane fixture recovered\n15, 7, and 0 reads. The two-group helper was exercised with 500 guides and two replicates per\ncondition in unpaired and paired modes; known depleted and enriched genes ranked first in their\nrespective directions and passed FDR 0.05. Sparse fixtures verified the total-normalization fallback.\nThese tests establish execution and signal direction, not real-screen statistical calibration.\nReal-file paths above are illustrative.\n\n`result\u002Freport.json` records count\u002Flibrary\u002Fcontrol-guide SHA-256, control-guide IDs, MAGeCK version,\nactual arguments, QC, applied normalization, warnings, hit direction, FDR, log2 fold change and rank.\nThe helper explicitly selects median guide LFC aggregation and `--remove-zero both`: all-zero\nguides remain in the input\u002FQC but are excluded from ranking. Native MAGeCK also skips `NA`\u002F`na`\ngene labels by default. The helper's `--fdr` filters completed gene results; it does not set MAGeCK's\n`--gene-test-fdr-threshold`, which controls the RRA guide-selection cutoff. Negative and positive\nFDRs are separate families; their union does not establish joint FDR control across directions or\nacross multiple contrasts. `screen.gene_summary.txt`, `screen.sgrna_summary.txt`, normalized\ncounts and the execution log retain the complete evidence. Include the original library, sample\nsheet and negative-control list in the analysis handoff; the count checksum cannot reconstruct them.\n\n## Primary references\n\n- [MAGeCK usage and command semantics](https:\u002F\u002Fsourceforge.net\u002Fp\u002Fmageck\u002Fwiki\u002Fusage\u002F)\n- [Official installation guidance](https:\u002F\u002Fsourceforge.net\u002Fp\u002Fmageck\u002Fwiki\u002Finstall\u002F)\n- [Input formats](https:\u002F\u002Fsourceforge.net\u002Fp\u002Fmageck\u002Fwiki\u002Finput\u002F)\n- [Output field meanings](https:\u002F\u002Fsourceforge.net\u002Fp\u002Fmageck\u002Fwiki\u002Foutput\u002F)\n",{"name":5,"license":14,"metadata":23,"description":7,"compatibility":28},{"version":24,"skill-author":25,"last-reviewed":26,"upstream-version":27},"1.1","K-Dense Inc.","2026-10-01","0.5.9.5","Requires Python 3.10+ for the helper and MAGeCK 0.5.9.5 with its compiled RRA executable for analysis. Tested runtime uses Python 3.11, NumPy 1.26.4 and SciPy 1.13.1. Source installation needs a C++ compiler; network is needed only for installation. No credentials.","main","66c9bd32ac14a921980362cf85872a1c402101c2","hermes\u002Fprofiles\u002Fworker-science\u002Fskills\u002Fmageck","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode","https:\u002F\u002Fgithub.com\u002Fbeanspowt","2026-11-10 23:22:33",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add beanspowt\u002Fbeancode --skill mageck",{"tool":41,"command":42},"bunx","bunx skills add beanspowt\u002Fbeancode --skill mageck",{"tool":44,"command":45},"pnpm","pnpm dlx skills add beanspowt\u002Fbeancode --skill mageck","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-science\u002Fskills\u002Fmageck\u002FSKILL.md",true,[49,55,59],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"references\u002Fdesign.md","REFERENCE","参考文档",2996,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-science\u002Fskills\u002Fmageck\u002Freferences\u002Fdesign.md",{"path":56,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":57,"sourceHtmlUrl":58},"references\u002Fruntime.md",3518,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-science\u002Fskills\u002Fmageck\u002Freferences\u002Fruntime.md",{"path":60,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":63,"sourceHtmlUrl":64},"scripts\u002Fscreen_analysis.py","SCRIPT","脚本",14068,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-science\u002Fskills\u002Fmageck\u002Fscripts\u002Fscreen_analysis.py",[66,73,80,87,94,101],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":18},"01M7F6BB3VS43C65P23JR4ZN5G","venue-templates","Prepares journal manuscripts, conference papers, research posters, and grant documents using venue-specific formatting guidance and bundled LaTeX scaffolds. Supports selecting an official template, checking current page or anonymity rules, adapting academic writing to a venue, or inspecting a submission PDF.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fvenue-templates\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":18},"01M7F6B4CBVEHY1RKVW55X8484","scientific-writing","Drafts, revises, and audits scientific manuscripts or reports with explicit evidence provenance, reporting-guideline coverage, authorship accountability, confidentiality controls, and local consistency checks. Use for manuscript sections, references, declarations, tables, figures, or submission preparation when scientific accuracy and traceability matter.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fscientific-writing\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":18},"01M7F6AXKSP5AXE8B7RCE1KP0C","scientific-visualization","Creates and audits truthful, accessible, publication-ready scientific figures with Matplotlib, Seaborn, or Plotly. Use it for figure design, multi-panel layouts, uncertainty and missing-data displays, color\u002Fcontrast review, image metadata validation, and journal export planning.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fscientific-visualization\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":18},"01M7F6AMRHS0QH64CMKE8TP9Y7","scientific-slides","Builds slide decks and presentations for research talks. Used for making PowerPoint slides, conference presentations, seminar talks, research presentations, thesis defense slides, or any scientific talk. Provides slide structure, design templates, timing guidance, and visual validation. Works with PowerPoint and LaTeX Beamer.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fscientific-slides\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":18},"01M7F6ACZ7GQJNNW0682235MC3","scientific-schematics","Generates scientific diagram drafts using Nano Banana 2 AI with smart iterative refinement. Uses Gemini 3.7 Flash for quality review. Refines when the review requests improvement, with at most two generations. Specialized in neural network architectures, system diagrams, flowcharts, biological pathways, and complex scientific visualizations.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fscientific-schematics\u002FSKILL.md",[],[],{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":18},"01M7F6A2JDHTYABYSA4NCE82VG","research-grants","Supports research proposal preparation and review for NSF, NIH, DOE, DARPA, and Taiwan NSTC, including opportunity-specific requirements, aims, review criteria, budgets, broader impacts, forms, and resubmissions. Use for investigator-authored grant development, compliance matrices, and proposal critiques.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fresearch-grants\u002FSKILL.md",[],[]]