academic-humanizer
Strips AI writing tells from academic papers and grant proposals.
A specialized editing skill for Claude Code, Codex, and MorphMind to refine academic drafts.
academic-humanizer removes generic AI phrasing from academic drafts and grant proposals while preserving scholarly voice and evidence. It is built for researchers and scientists who draft with AI but need to maintain strict academic precision.
What it is
It runs a six-layer editing process that trims fluff, matches claims strictly to evidence, and keeps citations, symbols, and numbers untouched. It includes a dedicated grant proposal mode optimized for the first-page structure of NSF and NIH applications.
Who it's for
Suits academic researchers and principal investigators who use LLMs for drafting papers or grant proposals and want to eliminate obvious AI tells. It does not suit bloggers, marketers, or anyone looking for tools to bypass AI detection or generate fake findings.
Key features
Trims AI Phrasing
Removes cliches like "In recent years", rule-of-three, and inflated phrasing.
Claim Evidence Matching
Ensures no verb is stronger than the data, changing words like prove to show empirically.
Scholarly Preservation
Keeps passive voice, evidence-tied hedging, equations, symbols, and every single citation intact.
Grant Proposal Mode
Focuses effort on the critical first pages of NSF and NIH proposals where reviewers score heaviest.
Up and running in 3 steps
Our honest verdict
What's great
- Calibrated against accepted papers and funded NSF and NIH proposals
- Keeps all citations, numbers, and references completely untouched
- Includes a dedicated mode for grant structure and first-page primacy
- Adapts directly to your personal writing style by pointing it at past papers
Where it falls short
- Requires an AI agent environment like Claude Code, Codex, or MorphMind to run
- Rules reflect one specific team's voice and require manual forking to fully adapt
- Does not check current binding requirements or page limits for funding agencies
Pricing & license
Free under the MIT license. The only cost is your own time and the API or compute costs associated with running your chosen AI coding agent.
Open remote AI-training roles
870 open remote AI-training roles, updated November 14, 2026; most pay $20–$170 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
Use it, or have us set it up
FAQ
Does this tool help bypass AI detectors?
No, it is explicitly not designed to evade AI-use detection or add fake novelty.
Does it change my citations or data?
No, it leaves every reference, number, and citation exactly as written.
What platforms does it support?
It runs as a skill or system prompt for Claude Code, Codex, and MorphMind.
