academic-humanizer is a skill that refines AI-assisted drafts of academic papers and grant proposals by sharpening clarity and matching claims to evidence while preserving the author's scholarly voice, numbers, and citations.
Project overview
Built specifically for academic precision rather than flattening text for blogs or marketing, it operates across a six-layer audit-rewrite loop that leaves real scholarship—including passive voice, definitions, symbols, and citations—untouched.
Project type
AI Agent · Prompt Engineering · Learning Resources
Use cases
Documents & Office · Learning & Education
Deployment
Refer to project documentation
License
License pending
Best for
Researchers and educators who need AI-assisted academic drafts refined for scholarly precision without losing their own voice, citations, or data.
Key capabilities
Trims generic AI phrasing such as 'paves the way' or 'In recent years...' and very long sentences, bringing drafts closer to the author's own style.
Keeps claims tied to evidence by limiting verb strength to data—for example, converting 'prove' to 'show empirically'—and converting vague magnitudes to attributed ranges.
Preserves evidence-tied hedging, passive voice, definitions, symbols, numbers, and citations without altering them.
A separate mode for NSF and NIH grant proposals that keeps the vision a paper-focused pass would trim and focuses effort on the first pages reviewers score.
Calibrates the draft to the author's prior accepted papers and a specified target venue.
Operates across six layers: general AI-tell catalog, academic-specific tells, scholarly conventions preservation, claim-evidence matching, voice and venue calibration, and funding-proposal mode.
Limitations and risks
The rules reflect one group's voice and must be forked and personalized for universal use.
The tool does not provide current binding requirements for page limits, formatting, or deadlines for NSF or NIH proposals.
Users might misuse the tool to evade AI-use detection despite explicit warnings against this purpose.
Getting started
Installation is a git clone command into the target agent's skill directory. After cloning, run the slash command and paste text or point at a file to begin.
GitHub project description: A skill for Claude Code, Codex, and MorphMind.
README: The problem is that AI-assisted drafts come out generic and verbose, with "In recent years..." openers, inflated phrasing, and over-long sentences. They also drift from the author…
README: We wanted AI-assisted drafts to read clearly and in the author's own voice, with the numbers, citations, and claims left exactly as written.
GitHub project description: Strip AI-writing tells from papers and grant proposals (NSF/NIH), while keeping scholarly voice and tying claims to evidence.