# Winston AI MCP MCP > Winston AI is a full-spectrum content integrity validator. It checks text for plagiarism against the web, measures authorship likelihood to detect AI generation, and verifies facts in articles or documents. This MCP gives you total visibility into whether any digital content—from student essays to news reports—is original, truthful, or synthetic. ## Overview - **Category:** artificial-intelligence - **Price:** Free - **Tags:** ai-detection, plagiarism-checker, content-integrity, text-analysis, academic-integrity, llm-verification ## Description Stop guessing if a piece of writing is real. Winston AI lets your agent validate the source, originality, and factual accuracy of any text, image, or document. Whether you're reviewing student submissions or vetting breaking news, this MCP provides concrete evidence about content integrity. Need to know if an article was written by a human or an LLM? Run `detect_ai_text` on the string directly, or upload a file for deep analysis with `detect_ai_file`. Is that claim actually true? Use the fact-checking tools to verify specific claims against trusted online sources. Did someone lift this entire section from another source? Check it with `check_plagiarism_url`, and you'll get a detailed similarity score. This suite handles everything: plagiarism, AI authorship, deepfakes, and factual errors. The power comes when you chain these checks together. You can connect your agent through Vinkius, and the platform's native token optimization cuts down on API costs while giving you full visibility into every step using Vinkius AI Analytics. Your workflow gets traceable, verifiable, and accurate. ## Tools ### check_plagiarism_file Checks a document file provided by URL for instances of plagiarism. ### check_plagiarism_text Analyzes pasted text to find sections that match existing online content. ### check_plagiarism_url Scans an entire website URL to determine its overall similarity score and potential for plagiarism. ### detect_ai_file Detects if a document file, accessed via URL, contains content likely generated by AI models. ### detect_ai_image Determines the likelihood of an image being AI-generated or manipulated using its URL. ### detect_ai_text Analyzes a string of text to calculate a score indicating how likely it is that a human wrote it, rather than an AI. ### detect_ai_url Scans a website URL to check if the content found there has been generated by artificial intelligence. ### fact_checker Verifies specific claims within a block of text against established online knowledge. ### fact_checker_file Checks factual accuracy across an entire document file using its URL for verification. ### fact_checker_url Compares statements on a live webpage URL to trusted sources to confirm accuracy. ### text_compare Compares two separate blocks of text and provides a percentage score showing how similar the language is. ## Prompt Examples **Prompt:** ``` Analyze this text to see if it was written by an AI: '[Paste text here...].' ``` **Response:** ``` I've analyzed the text. Winston AI returned a Human Score of 95%, meaning it is highly likely to have been written by a person. I also analyzed it sentence by sentence and found no significant AI patterns. ``` **Prompt:** ``` Check the article at 'https://techcrunch.com/2026/02/10/ai-new-models' for plagiarism. ``` **Response:** ``` Scanning the URL... Winston AI found a 12% similarity score. It identified matches in two other publications, which appear to be properly cited quotes. The rest of the content is original. ``` **Prompt:** ``` Is this image AI-generated? https://pbs.twimg.com/media/suspicious-photo.jpg ``` **Response:** ``` Detection complete! The image provided shows strong indicators of AI generation (Synthetic Score: 88%). Winston AI identified artifacts common in GAN-generated faces. It is likely a synthetic image. ``` ## Capabilities ### Detecting authorship Run `detect_ai_text` or `detect_ai_file` to determine the likelihood that content was generated by a person versus an AI model. ### Verifying originality Use `check_plagiarism_text`, `check_plagiarism_url`, or `check_plagiarism_file` to scan content and identify sources of copied material. ### Fact-checking claims Run fact verification tools against text, URLs, or files to check specific statements for accuracy using trusted external data. ### Analyzing image authenticity Determine if an uploaded image is synthetic or manipulated by running `detect_ai_image` via its URL. ### Comparing texts Perform detailed similarity analysis between two pieces of text using `text_compare` to spot paraphrasing and subtle changes. ## Use Cases ### Reviewing student theses for academic fraud A professor needs to check a stack of papers. Instead of manually checking each one, they use `check_plagiarism_file` on all submissions at once and run `detect_ai_file` on the most suspicious ones. They get a clear report showing both originality gaps and potential AI authorship. ### Fact-checking breaking news stories A journalist receives an article about local policy changes. They use `fact_checker_url` to verify every single date, name, and statistic in the piece against reliable government websites before publishing. This prevents retractions and saves credibility. ### Auditing client-submitted reports A corporate compliance officer receives a 50-page report from an external contractor. They run `check_plagiarism_file` to ensure the content isn't sourced elsewhere, and they use `detect_ai_text` on key executive summaries to flag potentially synthetic language. ### Vetting social media images A marketing team receives a viral image claiming something suspicious. They run `detect_ai_image` on the URL immediately; if it returns a high Synthetic Score, they know the content should be flagged as fake before anyone sees it. ## Benefits - Know if an essay passed through a bot. Use `detect_ai_text` to calculate the Human Score on any text string and verify authorship. - Stop accidental copyright infringement. Run `check_plagiarism_file` against large academic documents before submission deadlines. - Verify claims in real-time. Send a URL to `fact_checker_url` and get instant confirmation on whether the article's stated facts hold up to scrutiny. - Catch deepfakes immediately. Upload an image or provide a link, and `detect_ai_image` tells you if it was synthesized by AI. - Spot paraphrasing efforts. Use `text_compare` when reviewing drafts to see exactly how much two pieces of writing overlap in structure and phrasing. ## How It Works The bottom line is you get a single source of truth about the reliability and origin of any digital content. 1. First, connect your agent to the Winston AI MCP via any MCP-compatible client. 2. Next, provide the specific data you need analyzed (e.g., paste text into the prompt or supply a URL). 3. The tool returns scores and reports, telling you if content is plagiarized, how likely it is that an AI wrote it, or if facts are supported by online sources. ## Frequently Asked Questions **Is there a minimum text length for AI detection?** Yes, Winston AI typically requires at least 300 characters for an accurate assessment of whether the text is human or AI-generated. **Can I check if an image from a URL was created by AI?** Absolutely. Use the `detect_ai_image` tool and provide the public image URL. The AI will analyze it for signs of generation or manipulation. **How reliable is the fact-checking tool?** The tool cross-references claims against trusted online databases and sources to provide a verification score and identifying supporting evidence. **When I use detect_ai_text, how do I interpret the Human Score?** The score is a probability rating of human authorship. A high percentage means the text's patterns strongly align with typical human writing; conversely, low scores suggest significant AI influence. **Can check_plagiarism_url handle restricted or paywalled websites?** The tool checks publicly accessible content only. If a site requires login credentials or is behind a paywall, the crawler won't be able to access it for comparison. **What file types can detect_ai_file process for AI detection?** It supports common document formats like PDF and DOCX. Always ensure your uploaded file is in a standard, readable format; obscure or locked files won't work. **Does fact_checker verify claims against one database or multiple online sources?** It verifies claims using broad web indexing and trusted online data sources. This gives you cross-referenced context, making the results highly reliable for journalistic use. **Are there rate limits when I run text_compare on large batches of documents?** While we recommend batch processing through your agent client, API usage is subject to standard rate limits. You should monitor your usage dashboard for real-time quota information.