detect-ai

detect-ai is a skill for Claude Code from humanizerai/agent-skills. It costs 35 tokens per session (554 once invoked), scanned A, original, MIT.

A service-based checker that estimates whether supplied text was written by AI. It returns an overall score, a verdict, and supporting measures such as sentence length and word count.

In plain words
What is it for?
Use it to analyze text and report its AI-likelihood score, verdict, word count, sentence count, and related metrics.
Why use it?
It provides an automated review before publishing or submitting writing. The result can highlight text that may appear machine-generated, though it is an estimate.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: installed under .agents/ (shared by several agents).

Good fit Use it to analyze text and report its AI-likelihood score, verdict, word count, sentence count, and related metrics.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/humanizerai/agent-skills/detect-ai
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add humanizerai/agent-skills --skill detect-ai
Clone the repo
git clone --depth 1 https://github.com/humanizerai/agent-skills

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for detect-ai

README.md
[![agentmods](https://agentmods.dev/badge/skills/humanizerai/agent-skills/detect-ai.svg)](https://agentmods.dev/skills/humanizerai/agent-skills/detect-ai)
Your own site
<a href="https://agentmods.dev/skills/humanizerai/agent-skills/detect-ai"><img src="https://agentmods.dev/badge/skills/humanizerai/agent-skills/detect-ai.svg" alt="Measured on agentmods" height="20"></a>
Per session 35 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 554 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • Socket pass 18 Mar 2026
  • Snyk pass 15 Feb 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00035 $0.00554
Opus 5 $0.00017 $0.00277
Sonnet 5 $0.00007 $0.00111
Haiku 4.5 $0.00003 $0.00055

Measured 8d ago against content hash 4722632c7959, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

detect-ai scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agents/skills/detect-ai/SKILL.md · 90 lines

What it actually says

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:

{
  "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
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 90 lines · 35 tokens per session scan A 4722632c7959

Subscribe to this mod's changes

detect-ai is a skill published in the GitHub repository humanizerai/agent-skills (42 stars, last pushed 7mo ago), licensed MIT. It adds 35 tokens to every session and 554 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.