Claudeclear Mcp

io.github.aidevelopers2Published by GitHub account aidevelopers2v1.0.1Updated Oct 27, 2026

Remove the Claude text watermark by rewriting text, and strip C2PA/EXIF metadata from images.

VerifiedSTDIODesktop onlyMedia & DesignAI & ML

Overview

AI-generated overview

Rewrites Claude-generated text through a non-watermarking model and, in the local package, strips C2PA/EXIF metadata from images.

What you can do
  • Rewrite a passage of text (up to about 30,000 characters) through a different model and return the result
  • Strip C2PA and EXIF metadata from images (local stdio package only)
  • Report quota status when the monthly word allowance is used up
When to use it
Useful when you want AI-written passages reworded so their statistical watermark pattern no longer matches, or when you need image metadata removed locally. Not needed for ordinary editing or proofreading.
Requirements
Runs as a local stdio process via npx claudeclear-mcp, or as a remote HTTP connector at the vendor's endpoint with OAuth sign-in. The local package needs a ClaudeRewrite account and the CLAUDECLEAR_API_KEY environment variable; usage draws on the account's monthly word quota, with a small free allowance and paid plans.
Before you install
Text you submit is sent to the vendor's service along with your account identifier, and usage counts against your quota. The rewrite is not guaranteed to preserve facts, numbers, names or code, so review output before publishing. The API key is a secret tied to your account. The README's remote connector and the manifest's local package differ in tools and authentication.

Installation

In SourceWeft

  1. Open Claudeclear Mcp in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

ClaudeRewrite MCP connector

Rewrite Claude-generated text through a different, non-watermarking model so the statistical pattern in the wording no longer matches. Built by ClaudeRewrite.

One tool: rewrite_text(text). Sign in with your ClaudeRewrite account (OAuth) the first time you connect. There is no API key to copy.

Connect

Connector URL: https://removeclaudewatermark.com/api/mcp

Claude Code

bash
claude mcp add claudeclear --transport http https://removeclaudewatermark.com/api/mcp

Then run /mcp in Claude Code and choose Authenticate.

Claude Desktop (claude_desktop_config.json)

json
{  "mcpServers": {    "claudeclear": {      "type": "http",      "url": "https://removeclaudewatermark.com/api/mcp"    }  }}

Claude (web) and ChatGPT: add the URL as a custom connector, then sign in and choose Allow.

Full setup: https://removeclaudewatermark.com/docs/mcp

The tool

rewrite_text(text: string) rewrites a passage (up to about 30,000 characters per call) and returns the rewritten text. It is designed to keep facts, numbers, names and code intact; read the result before you publish it. It does not read your files, chats or history.

Accounts and limits

Usage comes out of your ClaudeRewrite monthly word quota, the same one the website uses. Free accounts get a small free monthly allowance; paid plans get more (pricing). When your quota is used up, the tool says so and nothing is rewritten.

Privacy and support

The connector receives only the text you ask it to rewrite and your account identifier. See the privacy policy. Questions: contact.

Other ways to use it

ClaudeRewrite is independent and is not affiliated with Anthropic or OpenAI.

License

MIT

Source: README.md at commit b169175

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v1.0.1LatestSep 16, 2026

Score

40% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

    Declare a license in the repository or the registry entry.