Winston AI for AI agents

Winston AI detects AI-generated text and plagiarism, used by educators and publishers checking whether content was written or copied rather than original. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.

AI and modelsAPI keyMCP + RESTgowinston.ai
AUDIT LOG · WINSTON AIPOLICY: acme-support
09:41:09 · claude · u_8f2write
winston_ai.text_compare
TEXT_COMPARE✓ approved · approved by user
EVERY WINSTON AI CALL, ON THE RECORD
01 · USE CASES

What agents do in Winston AI.

01

Detect AI-generated text

Check a piece of writing for signs it was generated by an AI model.

02

Check for plagiarism

Compare a document against known sources to flag copied content, a read-only check.

03

Compare two texts

Check how similar two pieces of text are to each other.

02 · ACTIONS

3 Winston AI actions, graded by risk.

Every Winston AI action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.

read

0

Look things up. Allowed by default.

  • No read actions.

write

3

Create and change things. Allow, or ask the user first.

  • winston_ai.text_compare
    TEXT_COMPARE
  • winston_ai.ai_text_detection
    AI Text Detection
  • winston_ai.plagiarism_detection
    Plagiarism Detection

destructive

0

Delete, cancel or archive. Ask first, or deny outright.

  • No destructive actions.
03 · HOW IT WORKS

Winston AI in three steps.

  1. 01Your users connect Winston AIThey add their Winston AI api key on Arc0 Connect, under your brand. It goes straight into the vault.
  2. 02You set the rulesReads run, and writes like “tEXT_COMPARE” can wait for the user to approve.
  3. 03Any agent can actYour agent calls Winston AI through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
POLICY.TS
await arc0.policies.set('winston_ai', {
  read: 'allow',
  write: 'ask',        // text_compare
  destructive: 'deny',  
})

# Claude Code: the same connection, one URL
$ claude mcp add --transport http arc0 \
    https://mcp.arc0.ai/u/u_8f2
04 · AUTH AND DATA

How Winston AI connects.

Users add their Winston AI api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.

The same Winston AI connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →

AUTH
API key
CREDENTIALS
Per-tenant encrypted vault
MODEL SEES
Results only, never credentials
AUDIT LOG
Every call, on every plan
05 · WORKS WITH

Use Winston AI from any agent.

Claude
ChatGPT
Cursor
Codex
VS Code
OpenAI Agents SDK
Claude Agent SDK
Vercel AI SDK
Mastra
LangGraph
07 · FAQ

Winston AI and Arc0, answered.

Q01

Can I use Winston AI with Claude, ChatGPT or Cursor?

Yes. Connect Winston AI to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Winston AI actions you allow.

Q02

How do users connect Winston AI?

Users add their Winston AI api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.

Q03

Which Winston AI actions can my agent take?

3 in total: 0 read, 3 write and 0 destructive, such as “tEXT_COMPARE”. Your policies decide which of them each agent may call.

Q04

Can I make my agent read-only in Winston AI?

Yes. Allow read actions and deny writes in the Winston AI policy. Your agent can still look things up, and any write it attempts is blocked and logged.

Q05

Can my own backend call Winston AI too?

Yes. The same Winston AI connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.

Get started

Plug Winston AI into your agent.

Your users connect Winston AI once, under your brand. Your agent gets 3 actions behind your policies, with every call on the record.

Free to build · MCP + REST · Audit log on every plan