AgentMint Verify

Verify agent payment endpoints before your AI spends

17 followers

Verify agent payment endpoints before your AI spends

17 followers

AgentMint Verify checks x402 and MPP payment challenges without a wallet or payment. It separates challenge, discovery, payment, settlement, delivery, and monitoring evidence into a reusable Trust Card. Run a free scan, use the REST or MCP endpoint, or start a seven-day drift monitor. The beta is read-only and sends no wallet credentials, signatures, or payment.

Free

Launch tags:API•Developer Tools•Artificial Intelligence

Launch Team

Maker

📌

I built AgentMint Verify after noticing that “supports agent payments” can mean very different things. An endpoint might return HTTP 402 but still lack machine-readable discovery, a verified settlement path, or proof that the paid response was delivered. The beta makes those claims explicit. Paste an HTTPS endpoint and it produces a six-layer Trust Card. The scanner is intentionally read-only: it sends no wallet credentials, signatures, or payment. You can also start a free seven-day monitor to catch protocol drift. The current beta charges nothing. The $29/month Founder tier shown on the site is an intent test, not an active checkout. I would especially value feedback from people building x402 or MPP endpoints: Which evidence layer is missing from your deployment workflow? Would a CI gate or hosted drift monitor be more useful? What proof would you need before trusting an agent to spend automatically?

Report

2mo ago

Maker

Maker update: we just shipped CI Gate + signed Watch webhooks after dogfooding the full production flow. If you maintain an x402/MPP endpoint, I'm looking for 5 developers willing to run a real URL through it and report false positives, missing schema variants, or setup friction. The beta is free and read-only: no wallet credentials, signatures, or payments; billing and email alerts remain disabled. Try it here: https://agentmint-verify.lasting... . Please reply with the protocol/version and what failed; don't share secrets.

Report

1mo ago

Forum Threads

p/agentmint-verify

•

2mo ago

What evidence should an AI agent require before paying an API?

HTTP 402 tells an agent that it must pay, but not whether pricing discovery, settlement verification, delivery, or retries work as expected.

While building AgentMint Verify, I chose to keep first scans read-only and separate each stage into its own evidence layer instead of producing one opaque safe/unsafe score.

View all