Detecting AI API and MCP documentation

Quickstart

Create an API key in your dashboard, choose an endpoint, and send JSON over HTTPS. REST requests use the public base URL below.

curl --request POST \
  --url https://api.detecting-ai.com/api/detect/ \
  --header "Content-Type: application/json" \
  --header "X-API-Key: YOUR_API_KEY" \
  --data '{
    "text": "Paste the text you want to analyze.",
    "version": "v3"
  }'

Authentication

Hosted MCP connectors use an automatic OAuth browser flow. REST integrations and local-config MCP clients can continue using a customer API key.

AI Detector API

Analyze supplied text for signals associated with AI-generated writing. Results are probabilistic and should support human review rather than replace it.

Request body

curl --request POST \
  --url https://api.detecting-ai.com/api/detect/ \
  --header "Content-Type: application/json" \
  --header "X-API-Key: YOUR_API_KEY" \
  --data '{
    "text": "Paste the text you want to analyze.",
    "version": "v3"
  }'

Response

{
  "success": true,
  "data": {
    "details": {
      "result": "Detector analysis is returned here"
    },
    "version": "v3",
    "words_processed": 8
  }
}

AI Humanizer API

Rewrite supplied text with a selected humanizer model while returning the processed word count and model name.

Request body

curl --request POST \
  --url https://api.detecting-ai.com/api/humanize/ \
  --header "Content-Type: application/json" \
  --header "X-API-Key: YOUR_API_KEY" \
  --data '{
    "text": "Paste the text you want to rewrite.",
    "model": "cognia"
  }'

Response

{
  "humanized_text": "The rewritten text is returned here.",
  "words_processed": 8,
  "model_used": "cognia"
}

Plagiarism Checker API

Submit text for plagiarism analysis. The response contains the checker result and the number of processed input words.

curl --request POST \
  --url https://api.detecting-ai.com/api/plagiarism/ \
  --header "Content-Type: application/json" \
  --header "X-API-Key: YOUR_API_KEY" \
  --data '{
    "text": "Paste the text you want to check."
  }'

Remote MCP server

Connect Detecting AI to Claude, ChatGPT, Codex, or another Streamable HTTP MCP client. MCP exposes focused detector and humanizer tools through the public endpoint below.

Claude and ChatGPT hosted connectors

Add the MCP URL in the client's connector settings. Enter only the URL: the client discovers Detecting AI OAuth, opens the sign-in and consent screens, and stores its own tokens. No API key needs to be copied into the connector.

1. In Claude, add a custom connector from Settings and Connectors. In ChatGPT, enable Developer Mode first, then add a custom connector.
2. Use https://api.detecting-ai.com/api/mcp as the connector URL.
3. Sign in to Detecting AI, review the requested permission and word-usage notice, then approve or decline the connection.

Clients with a local MCP config

Claude Code, Codex, and similar local clients can continue to use an API key stored in an environment variable or secret store.

Codex

# Set the key in your shell
export DETECTING_AI_API_KEY="YOUR_API_KEY"

# Add this to ~/.codex/config.toml
[mcp_servers.detecting_ai]
url = "https://api.detecting-ai.com/api/mcp"
bearer_token_env_var = "DETECTING_AI_API_KEY"
tool_timeout_sec = 180

Claude Code

claude mcp add --transport http --scope user \
  --header "Authorization: Bearer YOUR_API_KEY" \
  detecting-ai https://api.detecting-ai.com/api/mcp

Other local clients

{
  "mcpServers": {
    "detecting-ai": {
      "type": "streamable-http",
      "url": "https://api.detecting-ai.com/api/mcp",
      "headers": {
        "Authorization": "Bearer ${DETECTING_AI_API_KEY}"
      }
    }
  }
}

Errors

Usage and limits

Detector, humanizer, and plagiarism usage is measured from the input word count. MCP uses the same subscription balance as the REST API.

• Check remaining usage in the API dashboard.
• Avoid automatic retries after a successful response.
• Split very large documents into meaningful sections.
• Use a server-side timeout that fits the selected operation.