API Evangelist Paper · Market Report
The State of Education & EdTech APIs
The library out-publishes the classroom — scholarly infrastructure publishes a contract 70.8% of the time, corporate learning 0% — and seven organizations in 814 run an MCP server, the lowest of any market scored.
About this paper
Segment this market and the gradient runs precisely opposite to the money. The scholarly record and repository tier — non-profit, grant-funded, with no growth targets — scores 38.4 and publishes a machine-readable contract 70.8% of the time. LMS and learning platforms score 20.0 at 25.7%. K-12 student systems, which hold the personal data of tens of millions of children, score 17.3 at 15.9%. Corporate learning, the most commercial segment in the cohort, publishes none at all and returns an agent-readiness of 0.0.
The venture-funded end of this market is its least legible end, and the part with no capital in it is the one an integrator — or an agent — can actually use.
And this is the least agent-connected market measured. Seven organizations in 814 run a first-party MCP server: 0.9%, against a whole-catalog 5.7%. One reaches Agent-Native. One publishes idempotency. In the eighteen months that generative AI became the central question in every education institution on earth, that conversation has been almost entirely about students using models — and almost nothing has been built for models using the institution.
The third finding runs against the grain of the first two, and it is the one operators should read first: where this market does publish, it publishes carefully. Governance across the leaders is 48.8, the highest reading of any market scored this year, and rate-limit disclosure is 87.1%, also the highest. Developer ergonomics is 28.4, the lowest. This is a market that is good at the parts resembling publishing and weak at the parts resembling product — which makes its blueprint unusually cheap to work through.
What's inside
- Executive summary — the library out-publishes the classroom
- How this cohort was built, and why it matters
- The state of the market — distance from the classroom predicts the score
- The resource taxonomy — the vocabulary of the record, not the classroom
- The composite Kin Score, and why it's the foundation
- Agent-readiness — the market with the most at stake and the least built
- The blueprint
- The standards layer — settled specifications, unpublished implementations
- The regulatory layer — the strictest privacy rules, the least described consent
- Standards and artifacts for the agentic turn
- Security, auth, scopes and compliance
- Provider-by-provider intelligence
- The investable thesis
- Where this is going
database Includes the AI data bundle
This report ships with a machine-readable data bundle — the evidence behind every number, packaged to drop straight into the AI tool of your choice. Converse with the research you bought, check any claim against the source data, and take it further than the PDF.
folder_zip Delivered as a single ZIP alongside the PDF and Word edition at checkout.
What you get for $500.00
These papers are experience-based and vendor-neutral, distilled from the API Evangelist research at apievangelist.com. Questions before buying? [email protected].
arrow_back All papers