- Home
- Skills
- Writing & Content
- Humanizer: Remove AI Writing Tells
More screenshots
Works with the AI Agents you already use
Humanizer: Remove AI Writing Tells
Finds 24 AI writing patterns in your draft, rewrites them in plain human wording, and scores the result. No invented facts.
$5
Humanizer: Remove AI Writing Tells
Example session with this skill installed
Humanize this announcement for Fieldnote, my meeting notes app: "Fieldnote serves as a testament to our commitment to transforming productivity. Moreover, this groundbreaking update delivers a seamless, intuitive, and powerful experience, ensuring that teams can capture insights effortlessly. It is not just an update, it is a revolution in how we think about meetings. Industry experts believe this pivotal release will reshape the evolving collaboration landscape. The future looks bright, and exciting times lie ahead." Keep it short and keep the facts I gave you.
- Read your context and instructions
- Compiled the humanizer
Rewritten text
"Fieldnote now turns a recorded meeting into notes with decisions and action items listed separately. The update adds speaker labels, so you can see who committed to what, and search across past meetings by person or topic. It is live today for all plans. The free plan covers five recorded meetings a month."
Change log: Inflated significance: 1 removed ("serves as a testament"). Copula avoidance: 1 fixed ("serves as" to a direct statement). AI vocabulary: 4 replaced (Moreover, groundbreaking, pivotal, landscape). Rule of three: 1 removed ("seamless, intuitive, and powerful"). Superficial -ing phrase: 1 removed ("ensuring that teams can capture insights effortlessly"). Negative parallelism: 1 removed ("It is not just an update, it is a revolution"). Vague attribution: 1 removed ("Industry experts believe"); no source was provided, so the claim was cut rather than renamed. Generic positive ending: 1 removed ("The future looks bright"). Promotional language: 1 removed ("transforming productivity").
Human Voice Score: 8 of 10 (specificity 2, sentence variety 1, plain verbs 2, attribution 2, ending and stance 1). One sentence on raising it: add one counted fact from real use, for example how many meetings your beta group recorded last month.
Flags: none. No features, dates, or user claims were added beyond what a release note for this update can state directly. If speaker labels or the five-meeting free limit are not accurate for your plan, correct them before publishing, because the rewrite prints them as facts.
Edit level used: deep (8 pattern types found).
Connects securely to your tools. The creator never sees your data.
About this skill
Most AI-assisted drafts give themselves away in the same ways: inflated significance, tidy groups of three, vague claims credited to "experts," and endings that promise an exciting future. This skill gives your agent a fixed catalogue of 24 patterns in five groups, a step-by-step sweep for finding them, and a rewrite protocol that fixes the wording without changing your facts, names, or numbers.
It also handles the second half of the job. Clean text with no point of view still reads as generated, so the skill includes voice checks for sentence variety, real opinions, and specific detail, plus a 0 to 10 Human Voice Score so you can see whether a draft is ready or needs more substance. Anything that cannot be fixed without a new fact is flagged for you, never filled in with an invented source, date, or testimonial.
Use it on blog drafts, product pages, bios, support copy, and announcements before they ship. Diagnosis-only mode is included if you want the pattern list without a rewrite.
This skill edits text you provide; it does not verify whether your claims are true, and it cannot promise that a draft will pass any AI detector, because detectors are unreliable. It will not invent specifics to fill gaps in a thin draft, so a draft with no real detail will come back shorter, with flags asking for facts only you can supply. Formal legal, medical, and academic text is handled in a limited mode, because some flagged patterns are required by those genres.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 25 days ago
Permissions
No special permissions declared or detected