[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7F656TWRFX72Z0M5GFD7WVY":3,"skill-same:01M7F656TWRFX72Z0M5GFD7WVY":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},"01M7F656TWRFX72Z0M5GFD7WVY","relion",null,"Validates and executes RELION single-particle cryo-EM refinement and half-map postprocessing. Supports STAR optics\u002Facquisition checks, particle-stack consistency, gold-standard half sets, soft-mask validation, diagnostic Fourier shell correlation, and restart guidance.","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\u002Frelion\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: relion\ndescription: Validates and executes RELION single-particle cryo-EM refinement and half-map postprocessing. Supports STAR optics\u002Facquisition checks, particle-stack consistency, gold-standard half sets, soft-mask validation, diagnostic Fourier shell correlation, and restart guidance.\nlicense: MIT\ncompatibility: Python 3.12+ with numpy, mrcfile and starfile for bundled validation; RELION 5.0.1 CPU\u002FMPI executables for refinement and postprocessing. Native workflows require MPI, OpenMP and an FFT library (FFTW or MKL). GPU builds require their supported accelerator stack. Network is needed for installation only.\nmetadata:\n  version: \"1.1\"\n  skill-author: K-Dense Inc.\n  upstream-version: \"5.0.1\"\n  last-reviewed: \"2026-10-01\"\n---\n\n# RELION single-particle refinement\n\nUse for a RELION single-particle project, especially extracted particles → homogeneous selected\nparticle subset → gold-standard refinement → half-map validation and postprocessing. The bundled\nrunner starts from **CTF-annotated extracted particles and an initial 3D reference**. It does not\nreplace motion correction, picking, 2D\u002F3D selection, or a biological interpretation of map quality.\nFor tomography, helical reconstruction, Blush, or heterogeneous-state modeling, use the appropriate\nupstream workflow rather than forcing those data into this bounded SPA runner.\n\n## Preserve acquisition and coordinate conventions\n\nRead [references\u002Facquisition-and-restarts.md](references\u002Facquisition-and-restarts.md) when starting\nfrom movies or resuming jobs. Confirm pixel size in **Å\u002Fpixel**, voltage in **kV**, spherical\naberration in **mm**, defocus in **Å**, amplitude contrast as a fraction, and the symmetry justified\nby the specimen. Do not “correct” a suspicious value by guessing its units.\n\n`data_optics` describes acquisition\u002Fimage groups; `data_particles` references them through\n`_rlnOpticsGroup`. Particle filenames use **one-based** `index@stack.mrcs`; leading zeros such as\n`00000001@stack.mrcs` are valid. Relative paths resolve\nfrom the RELION project directory, not the STAR file's directory. Keep optics groups when merging\nor subsetting STAR files. `_rlnOriginXAngst`\u002F`_rlnOriginYAngst` are Å translations, not pixels.\n\nRun from this skill directory with paths to the real project:\n\n```bash\npython scripts\u002Fspa_workflow.py validate-star project\u002Fparticles.star --project project\n```\n\nThis opens referenced stacks and checks optics membership, finite acquisition\u002FCTF values, indices,\nbox sizes, duplicate particle references and existing half-set assignments. Use `--metadata-only`\nonly when stacks are genuinely unavailable; the JSON records `stack_checks_performed: false`.\nIt does not scan every particle pixel for corruption or establish correct image normalization.\nPhysical-range warnings are review prompts, not proof that unusual microscope settings are wrong.\n\n## Refine a selected particle population\n\nBefore running, inspect representative particles and class averages, defocus distributions, CTF\nfits, particle orientation distribution, and the initial reference. Ensure the map and particle\nboxes\u002Fpixel sizes agree after any downsampling. The runner deliberately supports one effective\nbox\u002Fpixel size across optics groups; handle heterogeneous sampling with an explicit upstream\nresampling workflow. Use conventionally extracted, normalized particles that have not already\nbeen phase-flipped or Wiener-filtered; this runner does not configure those special input cases.\n\n```bash\npython scripts\u002Fspa_workflow.py refine \\\n  --star project\u002Fparticles.star --reference project\u002Finitial.mrc \\\n  --project project --diameter 180 --symmetry C1 \\\n  --initial-lowpass 40 --mpi-ranks 3 --threads 2 --output project\u002FRefinePilot\n```\n\nThe diameter and low-pass filter above are illustrative **Å** values. Use specimen-appropriate\nvalues. Refinement executes `mpirun -np 3 relion_refine_mpi` with `--auto_refine`,\n`--split_random_halves`, `--ctf`, and a low-pass starting reference. Gold-standard splitting\nrequires MPI; the plain sequential `relion_refine` executable cannot perform this split.\nUse odd ranks ≥3 (master plus balanced half-set workers), with a matching MPI installation.\nThe CPU command is useful for a bounded pilot; choose a documented GPU\u002FMPI launch for full data.\n\nThe runner keeps the command, native version and log in a new output directory, records an\nexplicit random seed (default 1), surfaces runtime warnings, stops on process failure, and\nrequires converged unfiltered half maps before reporting success. It does not automatically retry\nexpensive jobs or silently discard failed-job artifacts. Keep `_optimiser.star`, model\u002Fsampling\nSTAR files, and referenced particle paths for restart. Use the original job's optimiser rather\nthan starting a new random split from a partially processed table.\n\n## Inspect independent half maps\n\nUse the two independently refined **unfiltered** half maps, never two copies of the combined,\nsharpened map. Matching headers cannot establish statistical independence; the independent\nparticle assignments and refinement history provide that evidence. Inspect directional\nanisotropy, preferred orientation and local resolution as well as a global FSC curve.\n\n```bash\npython scripts\u002Fspa_workflow.py fsc \\\n  project\u002FRefinePilot\u002Frun_half1_class001_unfil.mrc \\\n  project\u002FRefinePilot\u002Frun_half2_class001_unfil.mrc --output diagnostic-fsc.tsv\n```\n\nThis checks map dimensions, finite values, pixel size, origin, axis order and duplicate maps, then\nwrites an **unmasked diagnostic FSC**. The reported 0.143 crossing uses linear interpolation;\n`null` means no downward crossing was detected, not infinite resolution. Nyquist resolution is\n2 × pixel size. This diagnostic is limited to even cubic maps ≤256³; use RELION's native\n`relion_image_handler --fsc` for larger maps. It does not substitute for mask-corrected FSC.\nThe helper requires real-space maps with canonical axes, zero MRC start indices and orthogonal\ncell angles. Convert other grids explicitly with provenance; merely editing headers can misalign\ndensity. Matching headers and FSC cannot determine absolute handedness.\n\n## Postprocess with a soft mask\n\nConstruct the solvent mask from an appropriately low-pass-filtered density, with an expanded\nboundary and a smooth edge. Inspect all slices; a tight mask can inflate correlation. Avoid a\nmask derived from high-frequency noise shared between half maps.\n\n```bash\npython scripts\u002Fspa_workflow.py postprocess \\\n  --half1 project\u002FRefinePilot\u002Frun_half1_class001_unfil.mrc \\\n  --half2 project\u002FRefinePilot\u002Frun_half2_class001_unfil.mrc \\\n  --mask project\u002Fsoft_mask.mrc --output project\u002FPostProcessPilot\n```\n\nThe helper checks a nonconstant mask in [0,1], soft-edge voxels and matching map grids, then runs\n`relion_postprocess` with explicit half maps, mask and pixel size. RELION performs its own\nmask\u002Frandomization correction and writes `postprocess.star`. The bounded command leaves the\nB-factor at zero (no automatic B-factor estimation); add automatic\u002Fmanual sharpening only after choosing a defensible fit\nrange and inspecting map quality. A valid range and some fractional mask voxels do not prove the\nmask is scientifically appropriate. Inspect the phase-randomized masked FSC near the reported\nresolution: residual correlation calls for a smoother\u002Fwider mask and another postprocessing run.\n\nSee [references\u002Fruntime-and-validation.md](references\u002Fruntime-and-validation.md) for the tested\nnative utilities and the distinction between pipeline execution and reconstruction validation.\n\n## Primary references\n\n- [RELION 5.0 installation](https:\u002F\u002Frelion.readthedocs.io\u002Fen\u002Frelease-5.0\u002FInstallation.html).\n- [STAR and map conventions](https:\u002F\u002Frelion.readthedocs.io\u002Fen\u002Frelease-5.0\u002FReference\u002FConventions.html).\n- [Single-particle tutorial](https:\u002F\u002Frelion.readthedocs.io\u002Fen\u002Frelease-5.0\u002FSPA_tutorial\u002Findex.html).\n- [Gold-standard refinement](https:\u002F\u002Frelion.readthedocs.io\u002Fen\u002Frelease-5.0\u002FSPA_tutorial\u002FRefine3D.html).\n- [Mask creation and postprocessing](https:\u002F\u002Frelion.readthedocs.io\u002Fen\u002Frelease-5.0\u002FSPA_tutorial\u002FMask.html).\n","# RELION single-particle refinement\n\nUse for a RELION single-particle project, especially extracted particles → homogeneous selected\nparticle subset → gold-standard refinement → half-map validation and postprocessing. The bundled\nrunner starts from **CTF-annotated extracted particles and an initial 3D reference**. It does not\nreplace motion correction, picking, 2D\u002F3D selection, or a biological interpretation of map quality.\nFor tomography, helical reconstruction, Blush, or heterogeneous-state modeling, use the appropriate\nupstream workflow rather than forcing those data into this bounded SPA runner.\n\n## Preserve acquisition and coordinate conventions\n\nRead [references\u002Facquisition-and-restarts.md](references\u002Facquisition-and-restarts.md) when starting\nfrom movies or resuming jobs. Confirm pixel size in **Å\u002Fpixel**, voltage in **kV**, spherical\naberration in **mm**, defocus in **Å**, amplitude contrast as a fraction, and the symmetry justified\nby the specimen. Do not “correct” a suspicious value by guessing its units.\n\n`data_optics` describes acquisition\u002Fimage groups; `data_particles` references them through\n`_rlnOpticsGroup`. Particle filenames use **one-based** `index@stack.mrcs`; leading zeros such as\n`00000001@stack.mrcs` are valid. Relative paths resolve\nfrom the RELION project directory, not the STAR file's directory. Keep optics groups when merging\nor subsetting STAR files. `_rlnOriginXAngst`\u002F`_rlnOriginYAngst` are Å translations, not pixels.\n\nRun from this skill directory with paths to the real project:\n\n```bash\npython scripts\u002Fspa_workflow.py validate-star project\u002Fparticles.star --project project\n```\n\nThis opens referenced stacks and checks optics membership, finite acquisition\u002FCTF values, indices,\nbox sizes, duplicate particle references and existing half-set assignments. Use `--metadata-only`\nonly when stacks are genuinely unavailable; the JSON records `stack_checks_performed: false`.\nIt does not scan every particle pixel for corruption or establish correct image normalization.\nPhysical-range warnings are review prompts, not proof that unusual microscope settings are wrong.\n\n## Refine a selected particle population\n\nBefore running, inspect representative particles and class averages, defocus distributions, CTF\nfits, particle orientation distribution, and the initial reference. Ensure the map and particle\nboxes\u002Fpixel sizes agree after any downsampling. The runner deliberately supports one effective\nbox\u002Fpixel size across optics groups; handle heterogeneous sampling with an explicit upstream\nresampling workflow. Use conventionally extracted, normalized particles that have not already\nbeen phase-flipped or Wiener-filtered; this runner does not configure those special input cases.\n\n```bash\npython scripts\u002Fspa_workflow.py refine \\\n  --star project\u002Fparticles.star --reference project\u002Finitial.mrc \\\n  --project project --diameter 180 --symmetry C1 \\\n  --initial-lowpass 40 --mpi-ranks 3 --threads 2 --output project\u002FRefinePilot\n```\n\nThe diameter and low-pass filter above are illustrative **Å** values. Use specimen-appropriate\nvalues. Refinement executes `mpirun -np 3 relion_refine_mpi` with `--auto_refine`,\n`--split_random_halves`, `--ctf`, and a low-pass starting reference. Gold-standard splitting\nrequires MPI; the plain sequential `relion_refine` executable cannot perform this split.\nUse odd ranks ≥3 (master plus balanced half-set workers), with a matching MPI installation.\nThe CPU command is useful for a bounded pilot; choose a documented GPU\u002FMPI launch for full data.\n\nThe runner keeps the command, native version and log in a new output directory, records an\nexplicit random seed (default 1), surfaces runtime warnings, stops on process failure, and\nrequires converged unfiltered half maps before reporting success. It does not automatically retry\nexpensive jobs or silently discard failed-job artifacts. Keep `_optimiser.star`, model\u002Fsampling\nSTAR files, and referenced particle paths for restart. Use the original job's optimiser rather\nthan starting a new random split from a partially processed table.\n\n## Inspect independent half maps\n\nUse the two independently refined **unfiltered** half maps, never two copies of the combined,\nsharpened map. Matching headers cannot establish statistical independence; the independent\nparticle assignments and refinement history provide that evidence. Inspect directional\nanisotropy, preferred orientation and local resolution as well as a global FSC curve.\n\n```bash\npython scripts\u002Fspa_workflow.py fsc \\\n  project\u002FRefinePilot\u002Frun_half1_class001_unfil.mrc \\\n  project\u002FRefinePilot\u002Frun_half2_class001_unfil.mrc --output diagnostic-fsc.tsv\n```\n\nThis checks map dimensions, finite values, pixel size, origin, axis order and duplicate maps, then\nwrites an **unmasked diagnostic FSC**. The reported 0.143 crossing uses linear interpolation;\n`null` means no downward crossing was detected, not infinite resolution. Nyquist resolution is\n2 × pixel size. This diagnostic is limited to even cubic maps ≤256³; use RELION's native\n`relion_image_handler --fsc` for larger maps. It does not substitute for mask-corrected FSC.\nThe helper requires real-space maps with canonical axes, zero MRC start indices and orthogonal\ncell angles. Convert other grids explicitly with provenance; merely editing headers can misalign\ndensity. Matching headers and FSC cannot determine absolute handedness.\n\n## Postprocess with a soft mask\n\nConstruct the solvent mask from an appropriately low-pass-filtered density, with an expanded\nboundary and a smooth edge. Inspect all slices; a tight mask can inflate correlation. Avoid a\nmask derived from high-frequency noise shared between half maps.\n\n```bash\npython scripts\u002Fspa_workflow.py postprocess \\\n  --half1 project\u002FRefinePilot\u002Frun_half1_class001_unfil.mrc \\\n  --half2 project\u002FRefinePilot\u002Frun_half2_class001_unfil.mrc \\\n  --mask project\u002Fsoft_mask.mrc --output project\u002FPostProcessPilot\n```\n\nThe helper checks a nonconstant mask in [0,1], soft-edge voxels and matching map grids, then runs\n`relion_postprocess` with explicit half maps, mask and pixel size. RELION performs its own\nmask\u002Frandomization correction and writes `postprocess.star`. The bounded command leaves the\nB-factor at zero (no automatic B-factor estimation); add automatic\u002Fmanual sharpening only after choosing a defensible fit\nrange and inspecting map quality. A valid range and some fractional mask voxels do not prove the\nmask is scientifically appropriate. Inspect the phase-randomized masked FSC near the reported\nresolution: residual correlation calls for a smoother\u002Fwider mask and another postprocessing run.\n\nSee [references\u002Fruntime-and-validation.md](references\u002Fruntime-and-validation.md) for the tested\nnative utilities and the distinction between pipeline execution and reconstruction validation.\n\n## Primary references\n\n- [RELION 5.0 installation](https:\u002F\u002Frelion.readthedocs.io\u002Fen\u002Frelease-5.0\u002FInstallation.html).\n- [STAR and map conventions](https:\u002F\u002Frelion.readthedocs.io\u002Fen\u002Frelease-5.0\u002FReference\u002FConventions.html).\n- [Single-particle tutorial](https:\u002F\u002Frelion.readthedocs.io\u002Fen\u002Frelease-5.0\u002FSPA_tutorial\u002Findex.html).\n- [Gold-standard refinement](https:\u002F\u002Frelion.readthedocs.io\u002Fen\u002Frelease-5.0\u002FSPA_tutorial\u002FRefine3D.html).\n- [Mask creation and postprocessing](https:\u002F\u002Frelion.readthedocs.io\u002Fen\u002Frelease-5.0\u002FSPA_tutorial\u002FMask.html).\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","5.0.1","Python 3.12+ with numpy, mrcfile and starfile for bundled validation; RELION 5.0.1 CPU\u002FMPI executables for refinement and postprocessing. Native workflows require MPI, OpenMP and an FFT library (FFTW or MKL). GPU builds require their supported accelerator stack. Network is needed for installation only.","main","66c9bd32ac14a921980362cf85872a1c402101c2","hermes\u002Fprofiles\u002Fworker-science\u002Fskills\u002Frelion","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 relion",{"tool":41,"command":42},"bunx","bunx skills add beanspowt\u002Fbeancode --skill relion",{"tool":44,"command":45},"pnpm","pnpm dlx skills add beanspowt\u002Fbeancode --skill relion","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-science\u002Fskills\u002Frelion\u002FSKILL.md",true,[49,55,59],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"references\u002Facquisition-and-restarts.md","REFERENCE","参考文档",6113,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-science\u002Fskills\u002Frelion\u002Freferences\u002Facquisition-and-restarts.md",{"path":56,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":57,"sourceHtmlUrl":58},"references\u002Fruntime-and-validation.md",7103,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-science\u002Fskills\u002Frelion\u002Freferences\u002Fruntime-and-validation.md",{"path":60,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":63,"sourceHtmlUrl":64},"scripts\u002Fspa_workflow.py","SCRIPT","脚本",17110,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-science\u002Fskills\u002Frelion\u002Fscripts\u002Fspa_workflow.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",[],[]]