Webhook Catcher - The MCP-native webhook catcher for AI builders

Hey Product Hunt 👋 I built Webhook Catcher because webhooks kept making me feel stupid. Every time I wired up Stripe, Clerk, or GitHub, something wouldn't fire — and I could never tell whose fault it was. Did the provider not send it, or did my handler break? The only way to find out was to deploy a console.log, trigger the event again, and squint at the logs. A deploy per print statement. It's 2026 and debugging a webhook still felt like doing it blindfolded. So I built the tool I wanted: mint a URL in seconds (no sign-up), and watch every request land live — headers, payload, all of it. Replay any request to your localhost, edited or as-is. See exactly what Stripe actually sent, not what the docs claim it sends. For v2, I added the part I care about most: your AI agent can now read your webhooks. - If you write code: there's a token-scoped REST API, so you can pull captures straight into your scripts, tests, and CI. - If you build with an agent — Claude Code, Cursor, or you're a vibe coder who ships with AI and isn't totally sure what a webhook even is — there's a built-in MCP server. Connect it once and your agent reads the failing request itself and fixes your handler. You don't have to understand the payload. Your AI does. That last part is why I built v2. The people who struggle most with webhooks are often the ones building with AI for the first time — and they're exactly the ones who shouldn't have to become webhook experts just to ship a payment flow. It's free. No account, no pricing, no "premium beta." Secret headers are redacted before anything is stored, and captures auto-delete after 30 days. I'd love your feedback — especially: what's the webhook that's wasted the most of your life? — Anjuman