# MCP Quickstart > Zero to a first Copyleaks result inside your AI agent, in six steps. ## Before you start You need two things. **1. A Copyleaks account you can already scan with.** Sign in to Copyleaks and check you can run a scan there on a paid balance. A subscription, a prepaid balance, or an account your organization pays for all work. Free plans can't connect, and neither can an account whose plan has ended. A Copyleaks **API key doesn't count** - the API is a separate product, and having a key doesn't connect an agent. **2. An AI tool with a place to add a connector.** Claude, Codex, Gemini CLI, Antigravity, Cursor and VS Code all have a *Connectors* or *MCP* setting. A product without one can't connect, on any plan. Once connected, all twelve tools work in every client. Larger files and images need a client that can run commands on your computer - see [what works where](/mcp/overview#what-you-need-and-what-works-where). In Claude Desktop, that's the Code tab. --- ## Step 1 - Add Copyleaks to your agent The server is called `copyleaks_app`, and its address is `https://extensions.copyleaks.com/mcp`. ### Add it **Claude Code** - one line: ```bash claude mcp add --transport http copyleaks_app --scope user https://extensions.copyleaks.com/mcp ``` `--scope user` makes it available in every project on your machine. Then run `/mcp` inside a session to sign in. **Claude Desktop or claude.ai** - go to **Customize → Connectors**, press **+ → Add custom connector**, paste the address above, then **Add → Connect**. **Any other client** - copy its block from [Client setup](/mcp/client-setup). Blocks differ between clients. --- ## Step 2 - Approve what it can do Your browser opens. Sign in to Copyleaks, review what your agent is asking for, and approve - you land back in your connector settings or terminal. [Permissions](/mcp/permissions) explains each one. **Your agent can only do what you could do in Copyleaks.** It can't buy credits or change your plan, delete any scan, report, or profile, create share links, or see anything in another account. --- ## Step 3 - Start a new session Agents load connectors when a session starts, so one added mid-conversation won't show up yet. | Your agent | What to do | |---|---| | **Claude Desktop - Chat or Cowork tab** | **Quit the app completely and reopen it** - closing the window isn't enough. Then start a new chat or a new Cowork task | | **claude.ai** | Reload the page, then start a new chat | | **Claude Code** - terminal or the Code tab | Exit and relaunch. In an open session, `/mcp` picks it up without restarting | | **Codex** · **Antigravity** | Exit and relaunch the CLI | **In the Chat or Cowork tab**, open the tools menu in the message box and check Copyleaks is listed and enabled. Not there? Check you're signed in to the Claude account you added it on. --- ## Step 4 - Check it worked Ask your agent: > *"Check my Copyleaks credit balance."* A number back means you're connected. --- ## Step 5 - Run your first scan Paste a few paragraphs and ask: > *"Check this for plagiarism and AI."* Your agent submits the scan and checks back until it finishes. Ask how far along it is and you may get a percentage - never a finish time. What comes back: your **`Plagiarism score`** and your **`AI score`**, as percentages to one decimal place, plus **which checks actually ran**. **No `AI score`?** `AI Detection` needs at least **350 characters** - below that it doesn't run. A blank means *not measured*, not *nothing found*. ### Other ways to start a scan | Ask for | What happens | |---|---| | *"Scan this file"* | Including PDFs, scanned pages and photos - Copyleaks pulls the text out and scans that | | *"Scan these files"* | **Up to ten at a time.** Each is its own scan, run together rather than one after another | | *"Scan this page"* | A web address, as long as it opens without signing in | | *"Was this image AI-generated?"* | An image scan - about the picture, not the words in it. Needs a client that can run commands | | *"Run that scan again"* | Re-checks an existing scan against current sources | | *"…using my Client Reports profile"* | Runs it with the settings from a saved scan profile | --- ## Step 6 - Go deeper on the result This is the part that isn't in the report. Keep asking: > *"Which sources did I match?"* > *"Show me the exact text that overlapped with the third one."* > *"Walk me through everything flagged, in the order I wrote it."* Your agent starts with the summary and goes deeper when you ask. **To see the document itself, open the report.** Every finished scan comes with a link - `Open this scan in Copyleaks` - that opens your marked-up document. --- ## Worth knowing on day one - **It uses your normal credits** - the same balance as scans in the app, with nothing extra to buy. [Limits and access](/mcp/limits) - **Links that need a sign-in scan the login page.** Google Docs, SharePoint, Notion and the like return a sign-in page, and Copyleaks scans it and spends credits on it. Send the file instead. [Scanning](/mcp/scanning) - **10 scans a minute, on every paid plan.** Extra work waits, and your agent carries on by itself. - **Your own scans only.** Anything shared with you stays in Copyleaks. --- ## If something doesn't work **No Copyleaks tools?** Start a new session (step 3). Still missing? You haven't signed in - `claude mcp login copyleaks_app`, or `/mcp` in a session. Anything else: [Troubleshooting](/mcp/troubleshooting). --- ## Where to go next | | | |---|---| | **Everything you can ask for** | [Tools](/mcp/tools) - the twelve tools and what each one does | | **Credits, limits and access** | [Limits and access](/mcp/limits) - throughput, who can connect, and how to disconnect | | **Setup for your client** | [Client setup](/mcp/client-setup) - the exact block for each client | | **How results work** | [Results](/mcp/results) - scores, matched sources, AI content and AI phrases |