Agent-Readiness Assessment
A fixed-scope engagement that turns "are we agent-ready?" into a measured baseline and a sequenced plan — score your API surface against the open Agent Readiness framework, map it onto the standards onramp, and walk away with a prioritized roadmap and the descriptor that proves it.
A human developer can paper over a lot of API friction — an ambiguous error, an undocumented idempotency convention, a prose-only auth description, an HTML-only changelog. An autonomous agent cannot. Every implicit convention a human silently absorbs is a place an agent gets stuck, retries blindly, double-charges a card, or hallucinates a payload. Agent readiness is the discipline of removing those implicit conventions and replacing them with machine-readable signals.
Most teams know that matters, but they have no canonical definition of “agent-ready,” no sequenced path to get there, and no way to check hundreds of repositories against it. So “are we agent-ready?” stays an anxious, unbounded question. This engagement makes it a measured baseline and a plan — a fixed-scope assessment in four moves.
The engagement, in four moves
- Scan. I score your public API surface against my open Agent Readiness framework — a twelve-dimension model covering spec presence, auth-model clarity, idempotency, error semantics, rate-limit headers, dry-run, examples, MCP, and event contracts, plus the forward-looking signals of
/.well-known/api-catalog, machine-readable consent, and identified agent traffic. Each dimension is scored 0–3 from a real evidence URL and rolled up into an overall posture. - Map. I place the result on the Agent-Era API Standards Onramp — the sequenced ladder from OpenAPI through JSON Schema, Arazzo, discovery, MCP, agent skills, async, and agent-payment standards. This produces your canonical definition of agent-ready: where you are, what the next rung is, and what it unlocks.
- Gap. A prioritized findings list — minimum operational docs, an
AGENTS.mdwith explicit boundaries, AI-scrapable delivery, task-oriented MCP tools — highest-leverage fixes first, each backed by a specific evidence URL. - Onramp. A sequenced roadmap plus an API Onboarding Descriptor for your flagship APIs, so you leave with a repeatable pattern rather than a one-off cleanup.
Because the framework is public and I use it the same way across every provider on APIs.io, the number is neutral — not a badge I sell, but an honest outside read your teams and stakeholders can trust. It works as a one-time baseline, and it works even better as a before-and-after: score today, ship your improvements, and I re-score against the same rubric so you can prove exactly how far you moved.
What you walk away with
- A scored scorecard across all twelve dimensions, each backed by a specific evidence URL
- Your position on the standards onramp — the canonical, sequenced definition of agent-ready for your operation
- A prioritized remediation roadmap — what to fix, in what order, to move each dimension up
- An API Onboarding Descriptor for your flagship APIs, and a benchmark against reference and peer providers
- A re-assessment pass that quantifies your improvement as a defensible delta
Related reading
- The Agent-Era API Standards Onramp — the paper that scripts this engagement
- The Agent Readiness framework — the open twelve-dimension model, schema, and vocabulary
- API Evangelist Scores 83% Agent Ready
- Making the API Evangelist Network Agent-Readable
- Reviews — the broader honest outside read on your APIs and governance
Start with the research
Everything I know about what agent-ready actually means and how it is measured is already written down, priced, and yours to read tonight — no call, no scoping, no proposal. Start there.
-
The Agent-Era API Standards Onramp $25
The sequenced ladder from OpenAPI to agent payments — what rung you're on. -
Explicit Agent Boundaries $25
What an agent may do, written down where the agent can read it. -
Your Provider Checklist $1,500
Your own Kin Score and Agent Readiness against your peers, then every rubric check you are failing — what each is worth, what to do, and in what order. The only thing I write for the company it is about.
Free first: your Kin Score and Agent Readiness are already published on APIs.io. Look yourself up before you buy anything — the score costs nothing and it is the same rubric all of this research is built from.
If you would rather have this done with you than do it yourself, I take a small number of engagements a year — [email protected].
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.
