# Plagiarism Detector Pro [Health: Active]

**Category:** 🎓 Education  
**Repository:** https://github.com/scalesynthai/Plagiarism-Detector-Pro  
**GitHub Stars:** 1  
**npm Downloads (last month):** 1144  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/plagiarism-detector-pro

## Description
Offline text-similarity, citation and writing-pattern checks with a student writing coach, as a CLI and local stdio MCP server.

## Tools
Capabilities this server exposes over MCP:

- **plag_scan_text** — Measures lexical overlap and reports advisory writing-pattern diagnostics
- **plag_scan_file** — Ingests `.docx`, `.pdf`, `.tex`, `.ipynb`, `.md`, or `.txt` from disk
- **plag_academic_coach** — Scans unsupported claims, tone booster formal synonyms, and thesis score
- **plag_paraphrase** — Restructures overlapping text into 3 scholarly formulations with attribution
- **plag_generate_citation** — Resolves DOI / arXiv / Paper Title to BibTeX, APA, MLA, and IEEE
- **plag_alphabetize_references** — Auto-sorts and validates reference lists alphabetically
- **plag_compare_drafts** — Computes draft continuity %, added content %, and word delta

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "plagiarism-detector-pro": {
    "command": "npx",
    "args": ["-y","plagiarism-detector-pro"]
  }
}
```

## Documentation

## What Plagiarism Detector Pro does

The Plagiarism Detector Pro MCP server exposes academic writing utilities through the Model Context Protocol. Its core checks measure contiguous lexical overlap and produce advisory diagnostics about writing patterns. These results are intended for review and do not reproduce the private corpora or scoring systems used by commercial services such as Turnitin or Blackboard SafeAssign.

The server fits the 🎓 Education category and supports workflows for students, instructors, and researchers. An agent can inspect text supplied in a request, read supported documents from local storage, evaluate a thesis, generate citation formats, reorganize references, or compare two drafts. The available operations focus on analysis and review rather than submitting work to an external plagiarism database.

## How it works

Run the package in MCP mode with the `plag mcp` command. The MCP interface communicates through native JSON-RPC over a local stdio process. Text can be passed directly to `plag_scan_text`, while `plag_scan_file` reads a path on disk. File ingestion supports `.docx`, `.pdf`, `.tex`, `.ipynb`, `.md`, and `.txt` files.

The scanning tools accept options to exclude quotations and bibliographies. The academic coach checks for unsupported claims, formal “tone booster” vocabulary, and thesis strength. For overlapping prose, the paraphrase tool returns three scholarly restructuring options with attribution information. Citation generation accepts a DOI, arXiv identifier, or paper title and returns BibTeX, APA, MLA, and IEEE forms.

## Setup and configuration

Install the NPM package globally or invoke it through `npx`:

```bash
npm install -g plagiarism-detector-pro
plag mcp
```

For Claude Desktop, add a server entry that runs `npx -y plagiarism-detector-pro mcp`. The README provides configuration locations for macOS and Windows. Claude Code can register the same command with `claude mcp add plagiarism-detector-pro npx -y plagiarism-detector-pro mcp`.

The repository also documents Docker and Python setup for its web application, but those instructions are separate from the local MCP command. No MCP-specific environment variables are listed as required.

## Tools and capabilities

- `plag_scan_text` measures lexical overlap and reports advisory writing-pattern diagnostics.
- `plag_scan_file` reads supported document formats from a local path.
- `plag_academic_coach` reviews unsupported claims, formal vocabulary, and thesis score.
- `plag_paraphrase` creates three scholarly formulations of overlapping text with attribution.
- `plag_generate_citation` resolves DOI, arXiv, or paper-title queries into four citation formats.
- `plag_alphabetize_references` sorts and validates reference lists alphabetically.
- `plag_compare_drafts` reports continuity percentage, added-content percentage, and word-count change.

Together, these features make the Plagiarism Detector Pro MCP server suitable for pre-submission review, citation preparation, and revision analysis. The comparison tool is useful when an agent needs to describe how substantially a draft changed rather than only flagging matching phrases.

## Limitations and notes

The output is advisory. Lexical overlap and writing-pattern diagnostics should be reviewed by a person and should not be treated as an institutional plagiarism ruling. The project explicitly distinguishes its checks from proprietary systems and does not claim to reproduce their data or scoring methods.

The MCP tools listed here expose the documented seven operations; other CLI commands in the repository, including batch processing, certificate generation, cleanup, and combined checks, are not part of this MCP tool list. File-based analysis also requires the agent process to be able to access the supplied local path. The Plagiarism Detector Pro MCP server is therefore best suited to local document workflows where the relevant text or files are available to the MCP host.

_Full upstream README: https://allmcps.com/mcp/plagiarism-detector-pro/readme_