AI agent builders
Check what your AI agents and MCP servers expose, before you ship
AdminKit's free MCP check tells agent builders whether an endpoint is reachable and demands authentication, since public MCP servers left open without auth during prototyping can otherwise be called by anyone who finds them, not just the agent they were built for, and the Agent Trust API and a Netbase-run MCP gateway take it further.
Updated
At a glance
Part of the AI agents & MCP cluster — more tools, guides and terms on ai agents & mcp.
The problem
Model Context Protocol (MCP) servers are becoming a common way to give AI agents access to tools and data, but it’s easy to stand one up without authentication, especially during prototyping, and then forget to add it before the endpoint becomes reachable from the public internet. A public MCP server with no auth can be called by anyone who finds it, not just the agent it was built for.
The problem runs in both directions. Agents that call out to unfamiliar tools or endpoints generally have no built-in way to check whether that endpoint is what it claims to be before sending a request. And when a tool sends a response back, that response is just more text the model reads, which means it can carry prompt-injection content: instructions hidden in the data that try to steer the model’s behavior — a risk the MCP specification’s own security best practices call out directly, with nothing in most setups filtering for it today.
On top of that, many agent setups keep no audit log of what the agent actually called, when, or with what result, which makes it hard to reconstruct what happened after the fact if something goes wrong.
Talk to an expert
Huy Nguyen (David), Founder & CEO, Netbase JSC
Ask how AdminKit fits ai agent builders — setup, timeline, or pricing.
What can an agent builder check for free?
- Check if your MCP endpoint is reachable and whether it demands auth — a quick way to confirm your server isn’t open to the public without authentication.
For background on securing an MCP server more generally, see the how to secure an MCP server guide and the MCP developer reference.
How do you protect agents and MCP servers beyond a single check?
The Agent Trust API lets an agent ask a quick trust question about an unfamiliar MCP endpoint before calling it: GET /v1/trust/mcp?url= returns a 0-100 score, a verdict (trusted, caution, untrusted or unknown) and findings such as no authentication required, unsafe CORS or a high-risk tool exposed anonymously. It runs anonymously with an IP rate limit on api.adminkit.net.
For a server you run, Netbase deploys and operates the MCP & Agent Gateway in front of it as a custom engagement. It adds OAuth-based authentication without token passthrough, tool allowlists, per-caller rate limits, filtering of tool responses and an audit log of what was actually called — see a security gateway for MCP servers for the architecture. Talk to our team about your server, or start with the lighter MCP / agent hardening review on services.
Together the two cover both directions of the problem: the Gateway protects a server you run from the wrong caller and unfiltered responses, and the Trust API helps an agent you build decide whether an endpoint is worth calling in the first place.
This page is general guidance, not a security audit of your specific network — see our disclaimer for what AdminKit's guidance does and does not commit to.
Frequently asked questions
Can anyone call a public MCP server that has no authentication?
Yes, if it's reachable on the public internet and doesn't require auth, anyone who finds the endpoint can call it. The mcp-check tool shows whether a given endpoint demands authentication or not.
How does an agent know whether an unfamiliar tool or endpoint is trustworthy?
Mostly it doesn't; agents typically call whatever endpoint they're configured with, with no built-in way to check it first. AdminKit's Agent Trust API closes that gap for MCP: GET /v1/trust/mcp?url= scores an MCP endpoint 0-100 with a trusted, caution, untrusted or unknown verdict and the findings behind it.
What is prompt injection, and does it apply to tool responses?
Prompt injection is when content an AI model reads, including a tool or API response, contains instructions designed to manipulate the model's behavior. Responses from external tools are a realistic injection path if nothing filters them first.
How do we get the AdminKit MCP & Agent Gateway?
Netbase deploys and runs the gateway in front of your MCP server as a custom engagement — talk to our team at /contact/. On your own, the free mcp-check tool checks reachability and auth requirements for any MCP endpoint.
Does AdminKit log what my agent actually calls?
The AdminKit MCP & Agent Gateway, which Netbase runs for you, keeps an audit log of agent tool calls. The free check tool does not log calls; it only reports reachability and auth status for a single check.
More in the AI agents & MCP cluster
Want AdminKit set up for you?
Tell us about your setup — our team will scope AdminKit for it and, if you need it, add hands-on expert help from Netbase.
How AdminKit works with you
- Delivered remotely — Netbase JSC, the company behind AdminKit, has worked with clients in 80 countries.
- Contracts and invoices are issued by Netbase JSC; payment goes to the company by bank transfer to Netbase JSC's company account or PayPal Business.
- Talk to the CEO, Huy Nguyen (David), directly on WhatsApp or by email at [email protected].