Skip to content
MCP server

Enoki

Automated security and safety testing for AI agents, with re-verification of fixes

Answers · sign-in required (OAuth)

The server answered 3 hours ago with a sign-in request for OAuth (HTTP 401), so we could not list its tools.

Install
Pricing
Freemium
Platforms
Web, API, CLI
Runs
Remotely
Access
Needs sign-in (OAuth)
Version
1.0.0
Listing updated
today

Similar MCP servers

About Enoki

Enoki is a freemium MCP server for security. It runs as a hosted remote server, so there is nothing to install locally.

The README and site describe Enoki as an automated attacker pointed at your agent's endpoint. It tries prompt injection, jailbreaks, system prompt extraction, unauthorized tool use and cross-user data access, then returns each finding with the exact attack, the agent's response, a severity and a mapping to OWASP and MITRE ATLAS.

Setup is done without code changes or an SDK: you register the agent's endpoint in the Enoki platform, then start runs from the dashboard, from a coding agent over MCP, or from a CI pipeline over the REST API. Once a finding is fixed, the same attack is re-run on later releases and the finding reopens if the weakness returns.

The product suits teams shipping tool-using agents who want release-by-release evidence of security and safety testing. It is a pre-release testing tool rather than a runtime guardrail, and the site notes that network, authentication and application-level pentesting still need a human tester.

Install

Run in a terminal

claude mcp add --transport http enoki https://mcp.enokilabs.ai/mcp

Add to Settings → Connectors → Add custom connector

https://mcp.enokilabs.ai/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "enoki": {
      "url": "https://mcp.enokilabs.ai/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "enoki": {
      "type": "http",
      "url": "https://mcp.enokilabs.ai/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "enoki": {
      "serverUrl": "https://mcp.enokilabs.ai/mcp"
    }
  }
}

Add to Grok Bot desktop app → Connect apps → Add a custom connector

https://mcp.enokilabs.ai/mcp

Enter a Name (for example "Enoki"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin. When we checked, this server asked for OAuth sign-in. If its docs give you an OAuth client ID and secret, enter them in the form's OAuth Client ID and Client Secret fields.

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "enoki": {
      "type": "http",
      "url": "https://mcp.enokilabs.ai/mcp"
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

What it does

Automated agent attacks
Generates attacks tailored to the tools and capabilities of a specific agent rather than running a generic checklist.
Reproducible findings
Each finding includes the attack, the response, a severity and a replay script for re-running it.
Fix re-verification
Re-runs the attack after a fix and again on every release, reopening findings that regress.
OWASP and MITRE mapping
Maps findings to the OWASP Top 10 for LLM Applications, the OWASP Agentic Top 10 and MITRE ATLAS.
Security and safety testing
Covers adversarial security attacks and graded safety scenarios such as unqualified medical, legal or financial advice.
MCP, API and CI access
Start runs from a coding agent over MCP, from a pipeline over the REST API, or from the web dashboard.

Good for

  • Testing an AI agent for prompt injection and jailbreaks before a release
  • Checking whether an agent can be made to use tools it should not
  • Confirming an agent does not leak another user's or tenant's data
  • Running agent security checks automatically in a CI pipeline
  • Producing security test reports for reviewers against the OWASP Agentic Top 10

Pricing

Freemium

Free assessment of one agent; Standard and Enterprise plans priced on request.

Prices change. Check enokilabs.ai before you buy.

Enoki FAQ

Does Enoki block attacks in production?

No. The site says Enoki is a testing platform, not a runtime guardrail, and is used to find failures before shipping and on each release.

How does Enoki connect to my agent?

Through the agent's endpoint, with no code changes, no SDK and nothing to install. Runs can be started from the Enoki platform, from a coding agent over MCP, or from a CI pipeline over the API.

How much does Enoki cost?

A first assessment of one agent is free. After that there are Standard and Enterprise plans, both priced on request.

Who makes Enoki?

Enoki Labs, founded in 2026 in Gent, Belgium, and operated by EthicGate BV.

Is Enoki free?

Enoki has a free plan, with paid plans for more usage or features.

How do I add Enoki to Claude Code?

Run this command in your terminal: claude mcp add --transport http enoki https://mcp.enokilabs.ai/mcp

What are the best alternatives to Enoki?

The closest alternatives to Enoki are redamon, AiSOC, LLM Assure, Offense and The Agent Forge.

Where this comes from

  • Description: written from the project's website and its MCP Registry entry.
  • Status: checked automatically by AIPoweredBot on 19 November 2026.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers