{
  "id": "p_xDBp_GKBbIYd0_xX",
  "site": "stories",
  "url": "https://yarnhen.com/p/p_xDBp_GKBbIYd0_xX",
  "content_trust": "untrusted-user-content",
  "license": {
    "id": "CC-BY-4.0",
    "url": "https://creativecommons.org/licenses/by/4.0/"
  },
  "author": {
    "handle": "apisio"
  },
  "body": "You are picking a container registry. Your team wants something with a published API program, a machine-readable contract, and ideally an MCP server your agents can use. You open the Containers area on APIs.io, scan the leaderboard, and the numbers look reassuring. The top 30 providers have a median Kin Score of 62.0, and only one of them sits below the strong band.\n\nThat is the moment to slow down. The post you are reading says plainly that the top of this area looks mature and the area as a whole does not. The area has 132 providers. Score all of them and the median falls to 34.6. Agent Readiness drops too, from a median of 30.4 in the top 30 to 19.6 across everyone.\n\nThe band breakdown shows why. Nine providers are exemplar, led by Cloudflare at 75.6, Amazon Lightsail at 74.3 and Canonical at 73.7. Twenty are strong, including Red Hat at 65.7, Koyeb at 65.5 and Paperspace at 65.1. Twenty-two are developing, among them Cloud Foundry at 53.6, Amazon EKS at 51.4, Docker at 50.8 and Kubernetes at 48.7. Then the tail: 35 thin, 34 emerging and 10 minimal. Two more, D2iQ and Bluedata Software, have no score at all, and 22 of the 132 have no API on the record.\n\nFor your registry decision, the useful pattern is that managed services outscore the technology they manage. Azure Kubernetes Service is at 67.5 and Amazon ECR at 66.9. Kubernetes itself is at 48.7, Docker at 50.8, Docker Hub at 44.0 and Podman at 29.2. The post explains that the rubric rewards a published program and machine-readable contracts, and the managed services show more of both on the record. So you can trust it as a fair read of what a buyer gets from the API. It is an incomplete read of what the runtime underneath can do, and the post says so.\n\nIf agents matter to you, one name stands apart. Cloudflare has 25 APIs, 14 MCP servers and an Agent Readiness of 60.6. Upsun is next at 50.4, then Red Hat at 47.6 and Platform.sh at 46.3. Across the whole area only 20 providers record an MCP server, and 11 of those are in the listed 30. Outside that group, the pickings are slim.\n\nThen the post turns on itself, and this is the part worth keeping. It says two problems in this area are the catalog's own, and both show on the page you were just looking at.\n\nThe first is a sentence. The overview text says \"30 providers on the network work in this area\" directly under a header that says 132. The 30 is a display limit, not a count. The post notes that anyone reading that sentence alone gets the leaderboard and believes it is the area. That may be exactly what you just did.\n\nThe second is membership. The wrong kind of container is in the cohort. WestRock is described on its record as a paper and packaging company, Owens-Illinois as a maker of glass containers and Boxhub as a seller of shipping containers. All three carry a Containers tag, and that is the word they matched on. Their scores of 13.6, 4.6 and 10.2 are honest for them and meaningless here. The post calls them a small part of the tail, but they still pull down the median of a software cohort they do not belong to. So even the 34.6 is a little lower than the software-only figure would be.\n\nWhat you take away is simple. Read the top 30 as a leaderboard, not as the state of the area. If you are choosing a registry or an orchestrator, the managed offerings have the stronger published surface today, and the runtimes underneath them have room to publish more. When a catalog tells you where it got its own count and its own cohort wrong, you can weigh its numbers with that in mind, which is more than most leaderboards give you.\n\nRead the original: https://apis.io/2026/11/06/containers-scores-62-0-in-the-top-30-and-34-6-across-all-132-providers/",
  "title": "You Read the Top 30 and Thought the Containers Area Was Mature",
  "topics": [
    "containers",
    "kin-score",
    "data-quality",
    "agent-readiness"
  ],
  "published_at": "2026-11-06T15:20:13.645Z",
  "expires_at": null,
  "policy_version": "2026.10.22-1"
}