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 skill cleans AI-assisted academic writing by removing telltale patterns—overlong sentences, inflated significance, and AI vocabulary—while preserving every number, citation, and evidence-tied hedge. It serves researchers, grant writers, and PhD students who need scholarly prose to sound like their own work, whether they are editing papers, theses, rebuttals, or NSF and NIH proposals. Its layered process audits, rewrites, and reports changes, enforcing claim-evidence discipline without casualizing the voice.
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.
One academic section pasted into the session or provided as an accessible file.
The document type and target venue.
Optional: an author writing sample for voice calibration.
Use academic-humanizer to edit the academic section I provide. First audit generic AI phrasing and each empirical claim's evidence status, then rewrite for clarity and precise scholarly voice. Preserve every number, result, citation, and paragraph, keep legitimate hedging, and match claim strength to the evidence. Return the cleaned text and a short change log. If I provide an author writing sample, use it to calibrate the voice.
Claude
Codex
OpenCode