Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
This collection provides structured support for biomedical researchers navigating the transition from draft to submission. It focuses on the technical and ethical precision required for peer review, offering tools to calibrate claim strength and ensure internal consistency across complex documents.
The skills streamline the final stages of academic writing, helping authors identify placeholder data, remove LLM artifacts, and verify references to reduce moderation risks during preprint uploads.
WHO IT'S FOR
Biomedical researchers
preparing manuscripts for peer review and submission
Academic authors
responding to peer reviewer feedback and rebuttals
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Replace <input_file> with a CSV, TXT, or TSV dataset containing at least two usable numeric feature columns.
Replace <numeric_feature_columns> with the comma-separated feature columns, or request automatic numeric-column selection.
Replace <output_dir> with the desired result directory.
Rscript must be available, with the optparse and data.table R packages installed.
Make sure pca-dimensionality-reduction is installed in your agent before trying this task.
Use the pca-dimensionality-reduction skill to run a basic PCA on my tabular dataset at <input_file>. Treat rows as samples and use <numeric_feature_columns> as features. Save the explained variance, sample scores, feature loadings, and diagnostic figures under <output_dir>.
Replace <biomedical_topic> with a focused biomedical or biological topic, gene, pathway, disease, or methodology.
Live results require suitable API or RSS access; otherwise the skill must clearly label its output as knowledge synthesis.
Make sure preprint-surveillance-finder is installed in your agent before trying this task.
Use the preprint-surveillance-finder skill for a 14-day scan of <biomedical_topic> across bioRxiv, medRxiv, and arXiv q-bio. State whether you are using live retrieval or knowledge synthesis, summarize a few emerging topic clusters and quiet-but-notable areas, and provide manual search strings for verification. Do not invent papers, authors, or identifiers.
Claude
Codex
Cursor
OpenClaw
OpenCode