Newsletters
Two emails, both Monday mornings. Take either, or both.
API Evangelist
Highlights of the stories, conversations, and API knowledge published across the industry each week.
The Specification Layer
What actually merged into the thirty specifications the Linux Foundation stewards — OpenAPI, AsyncAPI, JSON Schema, and the rest.
About API Evangelist Papers
The short version. API Evangelist has been publishing white papers on the API space since 2010. A paper distills one practice into a single PDF of about twenty pages, and they are sold from papers.apievangelist.com.
The data underneath them is open source. If a paper is wrong, tell me and I will fix it in the next version. If it was not worth what you paid, tell me and I will refund it.
Fifteen years of white papers
I have been writing white papers on the API space for over fifteen years now — API management, security, design, discovery, governance, and much more. The API Evangelist Fundamentals papers are the continuation of that tradition.
The Fundamentals are not the lively, opinionated stories you get on the API Evangelist blog. They are true to white paper form — drier, down to business — while still bringing the API Evangelist expertise and flavor to a medium-form industry paper. One practice, distilled, in about twenty pages, for the operator who has to do the work on Monday.
How AI is used
I want to be straightforward about this. AI is part of how this research gets made, and the posture is the one spelled out in my stance on how I use AI — disclosure, not endorsement.
What sits underneath the writing is not a prompt. Over forty distinct artifact recipes — Agent Skills and pipeline steps — run against each individual API provider to search for, generate, or derive what that provider publishes: OpenAPI and AsyncAPI contracts, JSON Schema, scopes and security, plans and rate limits, MCP servers, agent skills, well-known files, and the rest. That evidence is what these papers are read from.
The Kin Score
Every provider in the catalog is scored on the Kin Score, a public rubric applied identically to everyone in it. Version 0.11.0 reads eight facets across 111 checks — discoverability, contract quality, governance, operational transparency, developer ergonomics, commercial clarity, a conditional regulatory layer covering eight industry regimes, and open source — and sorts providers into six bands. Alongside it, a standalone Agent Readiness rating reads fourteen dimensions of whether a machine could actually use what a provider ships.
The rubric is iterated on regularly, based on what provider profiling turns up. We repeat. We refine. We fix mistakes. We balance determinism and non-determinism wherever we can. When the rubric changes, providers are re-scored — which is why every paper carries a version number and a date, and why a number quoted from an older version may not match today's.
The data is open
All of the data compiled for this research is open source and published on the API Evangelist GitHub organization, and browsable at APIs.io. If you want to take a crack at making sense of an area or a market yourself, everything these papers are built from is there. What you are buying is the assembly, not access to the evidence. I think the approach baked into those artifact recipes and the Kin Score is a compelling one — but you do not have to take my word for it, because you can check it.
What you get, and what it costs
The papers are written for humans first. They are dense. They are technical, but they are not code. That density is necessary — there are a lot of moving parts in these markets — and every paper is distilled down into something anyone can read or scan.
- Papers — from $25, sold from papers.apievangelist.com. Instant download.
A human can buy with a credit card through Stripe checkout. A machine door is built and proven, so an agent will be able to purchase directly using the x402 protocol — it is not switched on for live purchases yet, and this page will say so when it is. The goal is to meet the market where it is today, while acknowledging that this varies from industry to industry and role to role.
If it is wrong, or if it was not worth it
- If a paper did not bring you the value you needed, tell me and I will refund your money. No form, no argument.
- If you find anything wrong or inconsistent, tell me and I will correct it in the next version. Corrections are the feedback loop this research runs on, and every provider is free to submit one or ask for a re-score at no charge.
Email [email protected] either way. A person answers.
Why any of this exists
These papers are for the operators, buyers and sellers doing the work. They are produced across a wide number of areas and industries using a consistent process, on Agent Skills and a Kin Score rubric that keep evolving from feedback — both human and agentic. What they are trying to do is combine sixteen years of API experience, technical and business, with fast-moving insight into an AI moment we are all still trying to make sense of.
Let me know how I can improve them.