Questions about GuardProxy
What is GuardProxy?
GuardProxy is a read-only API proxy service designed for AI agents that sits between an agent and a target API. It allows GET, HEAD, and OPTIONS methods to pass through while blocking POST, PUT, PATCH, and DELETE with a 403 response. The platform runs across more than 300 edge locations and offers a request inspector that logs method, path, upstream, reason, and timing without storing credentials or request bodies.
Who is GuardProxy designed for and how do I get started?
The service is intended for users who want to control agent traffic by allowing reads while stopping destructive writes at the edge. It is suitable for any stable HTTP API, including REST APIs, and requires no SDK to set up. Users can start by adding their API base URL, copying the generated proxy URL, and pointing their agent to it.
What are the pricing tiers and trial terms?
The pricing starts with a 14-day Pro trial that requires no card and no forced upgrade. The Free plan includes 20k requests per month, while the Recommended Pro plan costs $19 per month for 50k requests, and the Team plan is $49 per month for 250k requests.
How difficult is the setup and can I test it live?
Setup requires changing only a single URL to route traffic through the proxy, with no need for an agent, SDK, vault migration, or infrastructure project. Users can test the live proxy immediately to see how GET requests reach the upstream while write methods terminate.
What are the performance metrics and latency figures?
Typical overhead is less than 5ms, with specific latency figures of 3ms in US West, 5ms in Europe, and 4ms in APAC. The platform aims to add minimal delay while keeping protection at the edge.
What are the trial duration and cancellation policies?
The website states that every account starts with a 14-day Pro trial, and the proxy stays online on the Free plan. There is a cancel anytime policy, and no SDK is required for integration.
How can I verify the method blocking behavior?
The request inspector provides a live demo where users can choose a method to send to /anything to see how GET reaches the upstream while POST, PUT, and DELETE do not. This allows immediate verification of the boundary behavior.
How are credentials managed and stored?
The platform does not store API keys or request bodies, only processing authorization headers in memory. This design ensures that credentials remain local to the user's configuration and are not persisted on GuardProxy's servers.
How many edge locations are used and why is this important?
The service operates across more than 300 edge locations, providing a global network for policy checks. This distribution ensures that protection is applied close to the agents rather than routing all traffic through a single data center.
How can I contact GuardProxy?
You can reach GuardProxy by e-mail at [email protected].
What is the official website of GuardProxy?
The official website of GuardProxy is guardproxy.dev.
Does GuardProxy have reviews on Trustburn?
GuardProxy has no reviews on Trustburn yet. Anyone who has used it can write the first review on its Trustburn page.
Post your question publicly — other visitors, and whoever manages this page, can answer. It will be visible on this page.
Want to hear about this company again? Follow it and we’ll email you when a new review appears.
Based on reviews, listing data and the company's website. Verify anything important with the company directly.
Questions & Answers
How does the read-only approach protect AI agents from accidental data changes?
Trustburn Team suggests a starter question on new profiles — the business hasn’t answered yet.
Sign in to post
Your text is saved — you'll come right back to post.
Don't have an account? Create one
News & Press
No news yet. This is where the business shares its own updates and press.
Photos
No photos yet. Photos shared by reviewers will show up here.
About GuardProxy
GuardProxy provides a read-only API proxy service designed for use with AI agents. The service sits between an agent and a target API, allowing read methods such as GET, HEAD, and OPTIONS to pass through while blocking destructive methods like POST, PUT, PATCH, and DELETE with a 403 response. It offers a request inspector that logs method, path, upstream, reason, and timing information without storing credentials or request bodies. The platform runs across more than 300 edge locations, with authorization headers processed in memory rather than persisted to a database. Setup is positioned as requiring no SDK, with users changing a single URL to route traffic through the proxy.
- Website
- guardproxy.dev
- [email protected]
