Humanpen

Organization

humanpen

Humanize document files with HumanPen to lower their AI-detection score. Rewrite a whole .docx/.pptx, or only the passages a Turnitin/iThenticate AI report flagged, while keeping every fact, number, citation, table and formula intact. Also converts citations to one style (APA, MLA, IEEE, GB/T 7714), condenses a document to a word budget, and translates while preserving layout. Use when the user points at a document FILE (.docx/.pptx; translate also takes .pdf/.xlsx/.epub/.html/.txt) or such an AI report and wants any of these; not for text pasted into chat.

Overview

Publisher	humanpen
Repository	`humanpen-skill`
Skill name	`humanpen`
Stars	88
Forks	0
Bundled files	4
License	Apache-2.0
Links	Skill source Repository

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
4 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by humanpen on GitHub. Read the source before you install it.

Installation

Install the Humanpen AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/humanpen/humanpen-skill/tree/master
Enable the skill in any chat where you want it available.

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/humanpen/humanpen-skill.git \
  .claude/skills/humanpen

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Humanpen in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Humanpen on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Humanpen is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

HumanPen

A document-level AI humanizer. Point it at a document file and it uploads the file, waits for the job, and writes the result next to the source — you get a path back, never the document's contents. Formatting, tables, images, citations, and formulas survive intact.

Every job spends the account's credits. State the cost and get the user's agreement before running any job command. Per 1,000 words processed: humanize 100 credits, fix-citations 5, condense 20, translate 10, charged only on the words actually processed, with a 10-credit minimum per job; failed or cancelled jobs cost nothing.

Setup

bash
export HUMANPEN_API_KEY=hp_...   # create one at https://humanpen.net/settings/api-keys

New accounts get 100 free credits at https://humanpen.net. Without a key every command fails with instructions to get one.

Commands

Run from this skill's directory; python3 and nothing else is needed. --help on any command lists its flags. Add -o PATH to choose where the result lands (default <name>-polished.docx beside the source).

Humanize — the main operation

Lower AI-detection signals on a .docx / .pptx. Three ways to scope it:

bash
# 1. Whole document.  --strategy: balanced | aggressive
python3 scripts/humanpen.py humanize paper.docx --strategy balanced

# 2. Only the passages a Turnitin / iThenticate report flagged — the rest is left untouched
python3 scripts/humanpen.py humanize paper.docx --report turnitin.pdf

# 3. Continue a finished job for FREE (once per job): re-runs its still-flagged
#    passages against a fresh report. Pass the job_id that humanize returned.
python3 scripts/humanpen.py free-rehumanize <job_id> --report new-turnitin.pdf

Optional length control — EXPERIMENTAL, a word limit noticeably weakens AI-rate reduction, so skip it unless a length is required:

bash
python3 scripts/humanpen.py humanize paper.docx --min-words 1500 --max-words 1800
# Per-passage budgets for the report's flagged passages:
python3 scripts/humanpen.py humanize paper.docx --report turnitin.pdf --segments-file budgets.json

The other three operations

bash
# Convert every citation + the reference list to one style.  --style: apa7 |
#   mla9 | harvard | chicago_author_date | chicago_notes | ieee | vancouver |
#   gbt7714 | gbt7714_author_year | ama | acs | oscola
python3 scripts/humanpen.py fix-citations paper.docx --style apa7

# Shorten the whole document to a word budget
python3 scripts/humanpen.py condense paper.docx --max-words 4000

# Translate, keeping layout, tables and formulas.  --to: zh | zh-tw | en | ja |
#   ko | es | fr | pt | ru | de | pl | it   (--source-lang: same codes, default auto)
python3 scripts/humanpen.py translate paper.docx --to zh

Read a report · resume a job · check credits

bash
# A Turnitin / iThenticate report: overall AI %, and the flagged passages
python3 scripts/humanpen.py report turnitin.pdf

# Pick a running job back up (optionally download its result); show the balance
python3 scripts/humanpen.py status <job_id> --download paper.docx
python3 scripts/humanpen.py balance

Reading the output

A finished job prints JSON. Report its output (the result path) and credits_charged (the cost) to the user. Do not read the produced file back to summarise it unless asked — it is a full document, and the point of the path is that its contents stay out of the conversation.

report and a finished humanize print ai_percent as a number — or null when the report shows *. Turnitin prints * whenever AI writing is under 20%; it will not quantify that band, because of false positives. So null is usually good news, not a missing result. It is never zero: report it as "under 20%, which is where Turnitin stops printing a number", never "0% AI".

Worth knowing

Formats. humanize takes .docx / .pptx; fix-citations and condense take .docx; translate additionally takes .pdf, .xlsx, .pptx, .epub, .html and .txt.
Minutes, not seconds. The command waits and prints progress to stderr. Interrupting it does not cancel the job — status <job_id> picks it back up.
A second pass is the plan, not a failure — and it is free. If a detector still flags the result, free-rehumanize <job_id> with a fresh report rewrites only the still-flagged passages, at no credit cost (once per job, with a daily cap). Keep balanced; no need to escalate to aggressive.
Result files are kept 7 days, then deleted; the job record stays.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Humanpen AI skill do?

Why use Humanpen on TypingMind?

Because you install it once and use it with any model. Humanpen is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Humanpen in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/humanpen/humanpen-skill/tree/master. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Humanpen?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Humanpen?

As many as you like. As long as a model supports skills, you can use Humanpen with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Humanpen AI skill free?

Yes. It is published on GitHub by humanpen under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Pokemon Player

NousResearch

OrganizationPopular

Play Pokemon via headless emulator + RAM reads.

Docx

Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files) or Word templates (.dotx files). Triggers include: any mention of 'Word doc', 'word document',…

Openclaw Docker E2e Authoring

openclaw

OrganizationPopular

Author OpenClaw Docker E2E and live provider Docker lanes.

Openclaw Parallels Smoke

openclaw

OrganizationPopular

Prepare, snapshot, run, rerun, debug, or interpret OpenClaw Parallels guest install, onboarding, gateway smoke, and upgrade checks across macOS, Windows, and Linux.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

HumanPen

Setup

Commands

Humanize — the main operation

The other three operations

Read a report · resume a job · check credits

Reading the output

Worth knowing

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now