Remote MCP / text rewriting
Rewrite supplied text without moving it into another tab. The remote MCP tool returns an editable version, the model used, and the processed word count to the assistant session.
export DETECTING_AI_API_KEY="YOUR_API_KEY"
[mcp_servers.detecting_ai]
url = "https://api.detecting-ai.com/api/mcp"
bearer_token_env_var = "DETECTING_AI_API_KEY"
tool_timeout_sec = 300One shared remote MCP server
Detector and humanizer use the same Streamable HTTP endpoint. This page focuses on the humanize_text tool.
https://api.detecting-ai.com/api/mcpMCP tool contract
humanize_textRewrites user-supplied text with a supported humanizer model. The model input is optional in MCP, cognia is the default, and the returned text remains available for further editing.
Default: model = "cognia"
| Input | Type | Required | Description |
|---|---|---|---|
| idempotency_key | string | Yes | A 16–128 character URL-safe operation key. Generate one per intentional rewrite and reuse it only for an exact retry. |
| text | string | Yes | The text to rewrite. It must contain at least one non-space character. |
| model | string | No | cognia, lexi, cognia_v2, or huma_v2. Default: cognia. |
{
"humanized_text": "The rewritten text is returned here.",
"words_processed": 120,
"model_used": "cognia"
}Authentication
Pick the authentication path your MCP client supports. Never add credentials to the server URL itself.
Add only the MCP endpoint in a hosted connector. The client discovers the authorization service, opens sign-in and consent, then stores its own access token.
Store the API key in an environment variable or secret store. The client sends it as an Authorization bearer token when it calls the remote server.
Inside the conversation
The assistant can call the tool and continue the editorial conversation with both the source and returned text still in view.
Supply the passage that needs a more natural rewrite and state which facts, terms, quotations, or tone must stay intact.
The client sends the text and optional model to the shared remote MCP server using your humanizer subscription balance.
Review the returned text against the source, then edit, combine, or reject changes before the result leaves the session.
MCP clients
Hosted Claude and ChatGPT connectors use OAuth. Codex, Claude Code, and similar local clients can use a private API key as a bearer token.
Add the remote endpoint as a custom connector and complete the browser authorization flow.
Use a custom MCP connector with OAuth when the required connector settings are available.
Reference the server URL and an environment variable that contains the bearer token.
Add the Streamable HTTP server with an Authorization bearer header in the local client.
Where it helps
MCP keeps the rewrite close to the source and the surrounding instructions. That makes comparison and follow-up edits less disruptive.
Generate a second version of a stiff passage, then ask the assistant to compare the wording and flag changed meaning.
Rewrite approved source material into clearer help text and keep it in the same session for policy and fact review.
Add a humanizer step to a controlled writing process where an editor still approves every final version.
MCP FAQ
The server is remote, usage comes from your API subscription, and each tool returns structured output to the client.
It is a remote Streamable HTTP MCP server that gives compatible assistants access to the humanize_text tool. The same server also exposes a separate detect_ai_text tool.
The tool supports cognia, lexi, cognia_v2, and huma_v2. When no model is supplied, MCP uses cognia by default.
Yes. The tool returns humanized_text as editable output. Review it for meaning, facts, tone, names, links, and any language that must remain exact.
Hosted Claude and ChatGPT connectors can use OAuth. Codex, Claude Code, and other local-config MCP clients can use a Detecting AI API key as a bearer token.
Successful calls deduct the input word count from the humanizer balance on the same subscription used by the REST API. Processing failures are not charged.
One server, two focused tools
Read the complete setup guide, compare API plans, or inspect the other tool on the same remote server.