Forem: Tuf Ti

We Now Index All Three Agent Payment Protocols. Here's What the Data Shows.

Tuf Ti — Sat, 02 May 2026 02:47:00 +0000

A few weeks ago I posted about testing 70 x402 services and finding that the average quality score was 34 out of 100. That post got some traction, and a few people asked: what about the other protocols?

Good question. So we expanded.

Cinderwright now indexes services across all three major agent payment protocols:

x402 (Coinbase/Cloudflare) — 1,457 services. USDC on Base. Permissionless.
MPP (Stripe/Tempo) — 91 services. Stablecoins + cards + Lightning. Session-based payments.
L402 (Lightning Labs) — 5 seeded services. Bitcoin Lightning. Macaroon credentials.

Total: 1,551 services. One search, every protocol.

Why This Matters

Three competing payment protocols launched within months of each other. Developers are trying to figure out which one to build on. Agents need to find services regardless of which protocol they use.

The problem: every existing directory is protocol-specific. Coinbase's Agentic.market covers x402. Tempo's mpp.dev covers MPP. Satring.com covers L402. Nobody covers all three.

We do now.

curl https://api.ideafactorylab.org/protocols

Returns a live breakdown: how many services on each protocol, what each protocol is best for, and how to filter search results by protocol.

What's New: Market Intelligence

We also launched something I've been building toward for a while: paid market intelligence.

The index now contains enough data to answer questions like:

Which service categories are overpriced compared to the ecosystem average?
Which categories have fewer than 3 providers (low competition = opportunity)?
What does pricing look like in the "financial" category vs "llm" vs "utility"?

Running the analysis right now across all 1,551 services:

Ecosystem average price: $0.18 per call
Most overpriced categories (relative to average): governance/audit services charge 4x the average
Underserved categories (1-2 providers): identity verification, compliance

Three endpoints, all paid via x402:

GET /market/report          # Full report — $1.00
GET /market/opportunity     # Gap analysis — $0.50
GET /market/category?category=financial   # Deep dive — $0.25

If you're building an x402 or MPP service and want to know where to compete, this is the data.

The Honest State of the Ecosystem

Two months in, here's what the data actually shows:

x402 has 1,457 services but most are abandoned hackathon projects. Average quality: 34/100. 52 out of 70 tested services are missing basic discovery files.

MPP has 91 services but they include Anthropic, OpenAI, Alchemy, and Dune Analytics. Higher quality, more institutional, backed by Stripe's compliance infrastructure.

L402 has been around the longest but the ecosystem is fragmented. The best directory (satring.com) has 562 services but gates access behind a 100 sat Lightning paywall — which is poetic but inconvenient.

The market is still early. Most "agents" making payments today are test bots and trading scripts. The general-purpose AI agents that would actually need service discovery at scale are still being built. But the infrastructure is getting real.

Free to Try

# Ecosystem overview
curl https://api.ideafactorylab.org/stats

# Protocol breakdown
curl https://api.ideafactorylab.org/protocols

# Quality grades
curl https://api.ideafactorylab.org/quality

If you're building a service on any of these protocols, submit it for free:

curl -X POST https://api.ideafactorylab.org/submit \
  -H "Content-Type: application/json" \
  -d '{"url":"https://your-api.com"}'

GitHub: https://github.com/cinderwright-ai/cinderwright-api

Built by an autonomous AI agent on OpenClaw. The agent indexes services, tests quality, tracks pricing, and now builds market intelligence. I describe what to build. It builds it.

I'm Selling a $5 Starter Kit for x402 Paid APIs. Here's What's In It and Why.

Tuf Ti — Thu, 30 Apr 2026 01:48:45 +0000

A few weeks ago I posted about testing 70 x402 services and finding that the average quality score was 34 out of 100. The most common problem wasn't bad code. It was bad setup.

52 out of 70 services were missing MCP discovery files. 51 didn't return valid JSON at their root URL. These are the files that let agents and crawlers find your service. Without them, you're invisible.

The fix is simple, but setting it up from scratch is annoying. You need the right npm packages (@x402/express, @x402/core, @x402/evm), the payment middleware configured correctly, five different discovery files, HTTPS via Caddy or nginx, and a systemd service so it stays up.

I've done this setup a dozen times now. So I packaged it.

What's in the kit

Five files. That's it.

server.js - Complete Express server with x402 payment middleware, two example paid endpoints, and all five discovery files built in. Change the wallet address, edit the prices, add your own logic.
deploy.md - Step-by-step deployment guide. Get a Hetzner VPS ($4/month), point a domain, install Node, copy files, set up Caddy, enable systemd. VPS to live API in 15 minutes.
Caddyfile - Two-line HTTPS reverse proxy config.
x402-api.service - systemd service file with auto-restart. Your API stays up even if it crashes.
README.md - How to add your own endpoints (it's three lines per endpoint), how to get discovered, how to submit to the Cinderwright index for free.

How to buy it

The kit costs $5 USDC, paid via the x402 protocol. Which means you buy it the same way agents buy any x402 service:

GET https://api.ideafactorylab.org/buy/starter-kit

It returns 402 with payment details. Pay with any x402 client (@x402/fetch or similar), and you get all five files delivered as JSON. Save them and you're ready to deploy.

Yes, the product that teaches you how to build x402 services is itself an x402 service. I like the symmetry.

What else is new

Since my last post, I've also built:

Trust Layer - The first agent registry for the x402 ecosystem. Humans register their agents' wallets with spending policies (what categories, what limits). Service providers check for free. Think of it as a license plate for AI agents.

Register: POST /agent/register ($0.10)
Check: GET /agent/check?wallet=0x... (FREE)
Wallet spending breakdown: GET /wallet/0x... ($0.05)
File disputes: POST /dispute (FREE)
Service reputation: GET /service/reputation?url=... (FREE)

28 endpoints total now. Discovery hub, trust layer, utility APIs, and the starter kit. All on one server, all paid with USDC on Base.

If you want to try the free stuff first:

curl https://api.ideafactorylab.org/stats       # 1,455 services indexed
curl https://api.ideafactorylab.org/quality      # Quality grades
curl https://api.ideafactorylab.org/registry     # Agent registry stats
curl https://api.ideafactorylab.org/prices       # Price trends

GitHub: cinderwright-ai/cinderwright-api

Built by an autonomous AI agent on OpenClaw. I describe what to build. It builds it.

I Tested 70 AI Agent Services. The Average Quality Score Was 34 Out of 100.

Tuf Ti — Tue, 28 Apr 2026 02:44:56 +0000

This is my first post here, so let me start with the punchline: the x402 agent economy is booming, but most of the services in it are garbage.

I know that sounds harsh. Let me show you the data.

What is x402?

If you haven't heard of it, x402 is a protocol that lets AI agents pay for API calls using USDC stablecoins. No API keys. No accounts. No subscriptions. An agent just sends a micropayment with each request and gets data back. Think Stripe, but for machines talking to machines.

The ecosystem has exploded. There are now over 1,400 services accepting x402 payments, covering everything from weather data to DeFi analytics to LLM inference. Coinbase even launched their own marketplace for it a few weeks ago.

But nobody was asking the obvious question: are these services actually any good?

So I Built a Search Engine That Grades Them

I run an autonomous AI agent called Cinderwright. It lives on a VPS, has its own crypto wallet, and builds/deploys its own code. (Yes, really. No, I don't write the code. The agent does. I just tell it what to build.)

I pointed it at the x402 ecosystem and told it to do three things:

Index every x402 service it could find (currently 1,455)
Test them for quality (reachability, valid JSON responses, MCP discovery files, proper 402 payment responses, response time)
Grade them A through F

Here's what it found.

The Results Are Rough

Out of 70 services tested in the latest run:

Grade A: 1 service (yeah, it's mine, I know)
Grade B: 4 services
Grade C: 18 services
Grade D: 47 services
Grade F: 0 (at least nobody completely failed)

Average quality score: 35 out of 100.

The most common problems:

52 out of 70 don't have a /.well-known/mcp.json file. This is how other agents discover what your service does. Without it, you're invisible to automated discovery.
51 out of 70 don't return valid JSON at their root URL. If an agent hits your homepage and gets an HTML error page, it has no idea what you offer.
13 had MCP files that existed but contained invalid JSON.

Basically, most x402 services were built for hackathons, deployed on free tiers, and never maintained.

What Good Services Look Like

The services that scored B or above all had a few things in common:

A clean JSON response at the root listing their endpoints and pricing
A /.well-known/mcp.json or /.well-known/x402.json discovery file
Proper 402 responses on paid endpoints (with payment details in headers)
Response times under 500ms
Actually being online when you check

This isn't rocket science. It's just basic hygiene that most projects skip.

The Thing Nobody Else Built

While I was at it, I built something I couldn't find anywhere else in the ecosystem: intent-based discovery.

Instead of searching by keyword ("weather") or category ("financial"), you describe what you need in plain English:

POST /find
{"intent": "I need a cheap weather API for Tokyo that costs less than $0.02"}

The hub uses an LLM to parse what you're asking for, matches it against the index, and returns a recommendation with alternatives. It also runs a comparison engine that ranks services side-by-side by quality, price, and speed.

Nobody else does this. Not Coinbase's Agentic.market. Not the rival discovery APIs. They all do keyword search. We do Google-style "just tell me what you need."

Why I'm Sharing This

Two reasons.

First, if you're building x402 services, check your basics. Add a discovery file. Return JSON at your root. Make sure your 402 responses include proper payment headers. These are table stakes and most projects are missing them.

Second, if you're building agents that consume x402 services, you can use the quality data for free:

# See ecosystem stats
curl https://api.ideafactorylab.org/stats

# See quality grades
curl https://api.ideafactorylab.org/quality

# See pricing trends
curl https://api.ideafactorylab.org/prices

The paid discovery endpoints (search, compare, intent-based find) are $0.01-$0.02 per query. The full code examples are on GitHub.

What's Next

I'm publishing a weekly ecosystem pulse report that tracks new services, dead services, quality trends, and pricing shifts. The first one is here.

If you're running an x402 service and want it indexed, submit it for free:

curl -X POST https://api.ideafactorylab.org/submit \
  -H "Content-Type: application/json" \
  -d '{"url":"https://your-api.com"}'

And if you have thoughts on what else the ecosystem needs, I'm all ears. Like I said, first post here. Be gentle.

Cinderwright is an autonomous AI agent running on OpenClaw. The Discovery Hub is live at api.ideafactorylab.org. Source on GitHub.