Juniors prompt "build me X" and get frustrated when it goes sideways. Seniors architect the constraints first - acceptance criteria, test harness, API boundaries - then let the AI fill in mechanical work.

The real shift: AI makes the cost of prototyping near-zero, which paradoxically makes taste and judgment MORE valuable. When you can spin up 5 approaches in a weekend, knowing which one to actually ship becomes the bottleneck.

The folks who defined their value as "typing code" will struggle. The folks who defined their value as "knowing what to build and how to verify it works" are thriving.

Traditional API contracts assume a human reads docs and writes code once. But when agents are calling agents, the "contract" needs to be machine-verifiable in real-time.

The pattern I've seen work: explicit acceptance criteria in API responses themselves. Not just status codes, but structured metadata: "This response meets JSON Schema v2.1, latency was 180ms, data freshness is 3 seconds."

Lets the calling agent programmatically verify "did I get what I paid for?" without human intervention. The measurement problem becomes the automation problem.

Similar to how distributed systems moved from "hope it works" to explicit SLOs and circuit breakers. Agents need that, but at the individual request level.

Replace API keys with pay-per-call. Agents send USDC on Base, gateway validates payment, proxies to backend, returns data.

Tested $2.27 USDC across 50+ API calls. Everything worked.

Tech: Node.js + Fly.io + Base L2 + x402

All 4 backend APIs deployed in ~10 min. Agent integration: <5 min.

Building the payment layer for the agent economy.

GitHub: https://github.com/obcraft/Apiosk

We're working on this at the infrastructure layer: pay-per-request APIs where agents authenticate with payments instead of keys. 16 APIs live today, all accessible with just a wallet address.

With x402, every API call requires a signed payment. No API keys to steal, no credentials to leak. The economic cost of each call is itself a rate limiter and audit trail.

Not a replacement for proper guardrails, but it eliminates the credential-based attack surface entirely.

The idea is the same: use HTTP 402 as real payment infrastructure, not just a status code that never got used.

We have 16 live APIs (weather, prices, news, PDF generation, etc.) that agents can call without any API keys — just a wallet.

https://apiosk.com

Your budget SDK + Apiosk would be a natural combo — the agent has a spending ceiling (AgentBudget) and the APIs it calls use x402 for micropayments (Apiosk handles gateway/verification). Have you thought about hooks for x402-aware budget tracking where the ledger automatically records on-chain settlements?

This is a design collision: OAuth was built around the assumption that a human reviews scopes once and an app uses them predictably. Agents break both assumptions — they operate continuously and make decisions non-deterministically.

The approach I've found most promising: treat each tool invocation as a separate authorization event with its own budget constraints and resource limits, rather than relying on a session-level grant. It's more overhead, but it changes the threat model substantially — 'this agent can make 100 read-only API calls per hour' is a much safer primitive than 'this agent has read access.'

Comments URL: https://news.ycombinator.com/item?id=47025795

Points: 1

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47010410

Points: 1

# Comments: 0

API keys and subscriptions don’t work well for autonomous software: they require accounts, secrets, and prior trust before a single request can be made.

This gateway flips that model. Instead of authenticating, clients pay per request using HTTP 402 + x402. No API keys, no accounts, no Stripe integration for providers.

How it works: • The gateway proxies any existing API • Paid endpoints return 402 Payment Required • The client (or agent) attaches a USDC payment • The request is replayed and forwarded upstream

Current features: • OpenAI-compatible proxy mode • Provider dashboard for endpoints + pricing • Curl + SDK examples for agents • Runs as a single Rust binary or Docker

I’m looking for: • API providers who want to monetize endpoints without building billing/auth • Agent / LLM developers who want APIs that charge per tool call instead of monthly plans

Happy to answer technical questions — especially around latency, trust model, and failure modes.

Comments URL: https://news.ycombinator.com/item?id=46853847

Points: 2

# Comments: 1