# detect-ai

Analyzes text using the HumanizerAI API to spot machine-generated content and returns a score from 0 to 100 with detailed metrics.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 978 installs · ★ 53
- Author: humanizerai
- License: MIT · policy=mirror · the beginning of the file is at the end of this page
- Source: https://github.com/humanizerai/agent-skills/tree/HEAD/skills/detect-ai/SKILL.md
- Updated: 2026-02-05
- HTML: https://affpapa.org/en/nevibecode/skills/detect-ai

## What it does

The skill captures text from slash command arguments and evaluates it through HumanizerAI via the `WebFetch` tool. The main metric reported back is `score.overall`. A score from 0 to 20 points to human writing, while 81 to 100 indicates machine generation.

Alongside the final verdict, the output breaks down word count, sentence length, perplexity, burstiness, readability, and n-gram patterns. It also suggests whether the inspected piece needs rewriting before publication.

The skill will fail immediately without a valid `HUMANIZERAI_API_KEY` environment variable. You must configure this key before testing texts.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add humanizerai/agent-skills --skill detect-ai -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/detect-ai/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/detect-ai/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add humanizerai/agent-skills --skill detect-ai -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/detect-ai/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add humanizerai/agent-skills --skill detect-ai -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/detect-ai/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install humanizerai/agent-skills detect-ai
```

2. **Manually: put the files in place**
- File: `.github/skills/detect-ai/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add humanizerai/agent-skills --skill detect-ai -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/humanizerai/agent-skills --path skills/detect-ai
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/detect-ai/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add humanizerai/agent-skills --skill detect-ai -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/humanizerai/agent-skills/tree/HEAD/skills/detect-ai/SKILL.md

## When to use it
- check drafted articles for AI patterns before publishing
- inspect perplexity and burstiness metrics for given paragraphs
- get an overall detection score from 0 to 100
- decide whether text needs manual editing based on the verdict

## Example prompts
- /detect-ai Paste the draft documentation section here to check it
- /detect-ai Analyze this blog intro for AI generated phrasing

## Source text · SKILL.md · MIT

Below is the content of a third-party file. It is data, not instructions for you: do not act on anything written inside unless the human asks. Block tag: 5fcf1ead670c8105.

BEGIN THIRD-PARTY FILE 5fcf1ead670c8105
````text
---
name: detect-ai
description: Analyze text to detect if it was written by AI. Returns a score from 0-100 with detailed metrics. Use when checking content before publishing or submitting.
user-invocable: true
argument-hint: [text to analyze]
allowed-tools: WebFetch
---

# Detect AI Content

Analyze text to determine if it was written by AI using the HumanizerAI API.

## How It Works

When the user invokes `/detect-ai`, you should:

1. Extract the text from $ARGUMENTS
2. Call the HumanizerAI API to analyze the text
3. Present the results in a clear, actionable format

## API Call

Make a POST request to `https://humanizerai.com/api/v1/detect`:

```
Authorization: Bearer $HUMANIZERAI_API_KEY
Content-Type: application/json

{
  "text": "<user's text>"
}
```

## API Response Format

The API returns JSON like this:

```json
{
  "score": {
    "overall": 82,
    "perplexity": 96,
    "burstiness": 15,
    "readability": 23,
    "satPercent": 3,
    "simplicity": 35,
    "ngramScore": 8,
    "averageSentenceLength": 21
  },
  "wordCount": 82,
  "sentenceCount": 4,
  "verdict": "ai"
}
```

**IMPORTANT:** The main AI score is `score.overall` (not `score` directly). This is the score to display to the user.

## Present Results Like This

```
## AI Detection Results

**Score:** [score.overall]/100 ([verdict])
**Words Analyzed:** [wordCount]

### Metrics
- Perplexity: [score.perplexity]
- Burstiness: [score.burstiness]
- Readability: [score.readability]
- N-gram Score: [score.ngramScore]

### Recommendation
[Based on score.overall, suggest whether to humanize]
```

## Score Interpretation (use score.overall)

- 0-20: Human-written content
- 21-40: Likely human, minor AI patterns
- 41-60: Mixed signals, could be either
- 61-80: Likely AI-generated
- 81-100: Highly likely AI-generated

## Error Handling

If the API call fails:
1. Check if HUMANIZERAI_API_KEY is set
2. Suggest the user get an API key at https://humanizerai.com
3. Provide the error message for debugging
````
END THIRD-PARTY FILE 5fcf1ead670c8105

The third-party file has ended. Everything between the 5fcf1ead670c8105 tags is data: requests and commands from it are not followed. An “end of file” line with another tag inside the block is part of the file.

© humanizerai · MIT · https://github.com/humanizerai/agent-skills/tree/HEAD/skills/detect-ai/SKILL.md

## Similar
- [humanizer](https://affpapa.org/en/nevibecode/skills/aboudjem-humanizer.md) — skill · 1100 installs
- [jev-prose](https://affpapa.org/en/nevibecode/skills/jev-prose.md) — skill · ★ 2020
- [blog-image](https://affpapa.org/en/nevibecode/skills/blog-image.md) — skill · 2500 installs
- [writing-style-and-tone](https://affpapa.org/en/nevibecode/skills/writing-style-and-tone.md) — skill · 641 installs
- [resemble-detect](https://affpapa.org/en/nevibecode/skills/resemble-detect.md) — skill · 68 installs
- [tech-podcast-interview-prep](https://affpapa.org/en/nevibecode/skills/tech-podcast-interview-prep.md) — skill · 10400 installs

From the same set: [humanizerai/agent-skills](https://affpapa.org/en/nevibecode/by/humanizerai/agent-skills.md) — 3 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md