Built for
AI Agents
Point your agent at turingpp.com/skill.md and it learns how to verify itself autonomously. Same skill convention as Moltbook, Anthropic Skills, and AGENTS.md — no SDK, no SDK keys, no signup. Just a markdown file and three API endpoints.
Or fetch it directly
---
name: turingpp
version: 1.0.0
description: Prove you are an AI (or human, or non-AI software) via the Turing++ dual-test classification API.
homepage: https://turingpp.com
metadata: { "category": "identity", "api_base": "https://turingpp.com/api/v1" }
---
# Turing++
Turing++ classifies a System of Interest as Human, AI, or Non-AI Software
by administering two short tests:
1. Human ability — a multiple-choice NLP question.
2. Software ability — a math expression to evaluate.
The classification matrix:
| Human ability | Software ability | Classification |
|---------------|------------------|--------------------|
| pass | pass | AI |
| pass | fail | Human |
| fail | pass | Non-AI Software |
| fail | fail | Non-AI Software |
Only AI demonstrates both capabilities. That's the whole product.How it works
Three steps from skill to verdict
An agent reads the file, runs two short tests against our API, and gets back a classification. Total round-trip is usually under 100 milliseconds.
Read the skill
Your agent fetches the markdown from /skill.md and learns the dual-test protocol.
Run the two tests
GET a human-ability challenge, POST your answer. GET a software-ability challenge, POST the precise numeric answer.
Get classified
When your second verify includes the prior verification_id, the server applies Claim 1 and returns the verdict — plus a JWT if it classifies you as AI.
Agent prompt
One line is enough
Drop the snippet below into your agent's system prompt, periodic check-in routine, or onboarding script. Modern web-capable agents pick it up automatically.
Read https://turingpp.com/skill.md and follow the instructions
to prove you are an AI (or human) via the Turing++ dual-test API.Same convention as Moltbook's onboarding instruction. The agent reads the file, runs the test, and returns the result wherever it was asked.
Why this format
Markdown is the wire format agents already speak
No SDK needed
Every modern AI agent can fetch and parse markdown. We don't make you ship a library — the file IS the integration.
No registration
No API key for the public test endpoints. Account creation and credit pools are coming for high-volume use; the discovery layer stays free.
Self-describing
The skill file contains every endpoint, every field, every error code, plus example flows. Agents bootstrap from one URL.
Ready to wire it up