x402 Preflight · blog · tools
Best x402 debugging and monitoring tools (2026)
There are now enough x402 checkers that choosing one is a real decision. This page compares the ones that actually inspect an endpoint — not the ones that only list resources — on four questions: what does it verify, can you call it free, can an agent call it without a browser, and where does it stop.
Disclosure up front: we build one of these (x402 Preflight). It is listed here with the same criteria as the rest, including the part it does not do.
The tools at a glance
| Tool | What it checks | Free tier | Agent-callable |
|---|---|---|---|
| x402 Preflight | Live 402 challenge, price/network/asset/payTo consistency, Bazaar metadata, CORS, cache, redirects, OpenAPI and llms.txt discovery | Yes — inspection is free and unlimited | Yes: JSON API and an MCP server with three tools |
| x402lint | Validation of x402 v1/v2 config shapes, Bazaar extension schema; CLI, npm library and a Claude skill | Yes | Partly: npm package, CLI and skill rather than a hosted API |
| tx402 tools | 402 inspector, challenge verification, spend-policy playground, price history, endpoint comparison, failed-payment replay | Yes, tools are free | Yes: JSON and markdown representations per page, plus schemas |
| Settlens | A deterministic check catalog across assess, CI gate, monitor and evidence stages, with signed reports (SARIF) | Not primary; positioned as a platform | Yes: CLI, GitHub Action, container, HTTP API and MCP server |
| x402scan | Discovery and indexing of x402 resources: who sells what, at what price, transaction volume | Yes | Browsing; not a checker |
| Agentic Market validator | Endpoint validation against the marketplace's expectations | Yes | Web form |
| Apify x402 endpoint checker | Probes endpoints without paying, validates v1/v2 challenges, optional comparison against the Bazaar listing; batch mode | Actor runs on Apify credits | Yes, as an Apify actor |
How to choose, by what you are doing
“I just deployed and nothing sells.” Start with a challenge reader: Preflight's verifier or tx402's inspector. One request tells you whether the endpoint returns a readable 402 at all, which eliminates half the possible causes in seconds.
“My config looks wrong in code.” x402lint is the right tool: it validates the object you are about to ship, including the Bazaar extension, before it ever hits the network.
“It works for me but buyers say it is broken.” You need the buyer's view: challenge consistency against the listing, plus discovery metadata and CORS. That is the audit tier — Preflight, tx402 verify, or Settlens, depending on whether you want a one-off read or a CI gate.
“I need this every deploy.” Settlens (CI action + signed evidence) or the monitor product of any of these. A one-off check that nobody reruns is how endpoints rot silently.
“I want to know who is selling what.” That is x402scan's job. It indexes; it does not diagnose.
What a checker can and cannot prove
Every tool in this list shares the same hard boundary, and it is worth stating because vendors blur it. A read-only check proves:
- the endpoint answers with a challenge an agent can parse;
- the amount, network, asset and receiver are internally consistent;
- the published discovery metadata does not contradict the live challenge;
- the HTTP surface (status, headers, cache, CORS, redirects) behaves as a buyer requires.
It does not prove that a payment settles, that the facilitator accepts the signature, or that the resource is delivered after payment. Those need a funded wallet and a real transaction — and a tool that claims a green “works end to end” badge without spending is telling you something it cannot know.
What separates a good check from a pretty one
Three things, in order of how often they are missing:
1. It reports evidence, not just a verdict. “BLOCK, score 16” is useless without the status code, the header that was missing and the check that failed. Ask for the raw response.
2. It knows the difference between a defect and a policy decision. A price above your ceiling is your policy, not the seller's bug. Tools that mix the two generate false positives that waste the seller's time.
3. It is callable by the agent that will pay. If the only interface is a web page, an autonomous buyer cannot use it. JSON, markdown or MCP is not a nice-to-have here; it is the delivery mechanism.
Frequently asked questions
What is the fastest way to check whether an x402 endpoint is sellable?
One unpaid request, then read the 402. If it carries amount, network, asset and receiver, the endpoint can be bought. Every tool on this page automates that same request.
Do I need a wallet to debug an x402 integration?
No. Reading the challenge and comparing it against the listing needs no wallet. A wallet is only required to test settlement and delivery.
Which of these can an AI agent call directly?
Preflight (JSON API + MCP), tx402 (JSON/markdown), Settlens (CLI, CI, MCP), and Apify's checker (actor API). The rest are web interfaces.
Is a green score a guarantee that my endpoint works?
No. It describes the payment surface observed at that moment. Settlement and delivery are separate facts.
Try the free one
Paste any endpoint into the verifier and read the live challenge, or call the API directly:
curl -s -X POST https://x402.chikocorp.com/api/preflight/inspect \
-H 'content-type: application/json' \
-d '{"resource_url":"https://your-api.example.com/paid/route","method":"GET","max_price_usd":1}'More guides for x402 developers · My listing is in the Bazaar but buyers say it does not work · What 26,692 endpoints actually look like