You are about to point an agent at Ubuntu's infrastructure. You want it to talk to the Snap Store, maybe spin up something in LXD, and you have heard Canonical has one of the widest public API estates around. So you open the APIs.io profile for Canonical and look at the number first: 74.8 on the Kin Score, exemplar band.
That looks like a green light. Then you read the next line. The score is falling, down 7.0 from 81.8 at the previous run. And Agent Readiness sits at 43.3, agent-ready, with a note that a gate held Canonical back from agent-native.
You keep reading, because the width is real. Canonical's provider record says 13 APIs on a published basis: the Snap Store, Charmhub, MAAS, Juju, Launchpad, Ubuntu Pro, LXD, Landscape, Pebble, MicroCeph, the Ubuntu Security API, Testflinger and the Anbox Cloud surface. The catalog holds 135 API pages under the provider, and the score counts 117 contracts. Most of those pages are single-resource slices with names like Canonical Network Zones API, Canonical Placement Groups API and Canonical Sentry Debug API.
Here is the part that changes your plan. Of those 117 contracts, 12.0% are callable from outside. That is on purpose, for much of the estate. LXD, MAAS, Pebble, Landscape and MicroCeph run on a customer's own machine, socket or cluster. The MAAS page says so honestly, with a base URL of https://<maas-host>/MAAS/api/2.0. You would never expect to hit those from the open internet.
Now the turn, and it is aimed at the catalog rather than at Canonical. 116 of the 135 API pages show https://api.snapcraft.io as their base URL, including the network, cluster and instance slices that belong to LXD. The post is blunt about whose fault that is: it is the catalog's defect, not Canonical's. If you let an agent trust those pages, it would send LXD calls to the Snap Store. So the first thing you learn is not to take the base URLs on those slices at face value.
The facet scores tell you where Canonical is strong and where it is not. Access clarity is the best at 92.1, and developer ergonomics is 73.2. Most of these APIs are open, documented and sit next to their source code on GitHub, in repositories such as canonical/lxd, canonical/pebble, canonical/testflinger and canonical/identity-platform-api. Discoverability is 69.6, contract quality 66.6 and operational transparency 60.5. Contract governance is the weakest at 35.6, with marker coverage of 2.6%, meaning almost none of the 117 contracts carry the versioning, lifecycle and ownership markers the rubric looks for. Under the Horizontal regulatory regime Canonical scores 45.1.
For your agent, the readiness record is mixed. On the good side, error semantics are verified, and idempotency, reversibility, rate-limit signalling and delegated identity are documented. Launchpad publishes a webhook surface. On the other side, there is no MCP server, no agent card, no well-known catalog and no protected resource metadata. The agentic-access artifact, at 812 operations with 413 acting and 3 human-in-the-loop, and the Agent Skills dimension are both marked derived. The catalog built them, so you should not credit them to Canonical.
One more caution before you repeat the seven-point drop to anyone. The record shows the fall but not its cause. Before you call it a regression at Canonical, the post says it should be checked against the rubric version that scored the previous run.
What you take away is practical. Treat the 74.8 as a verdict on documentation and openness, which are already exemplar, and not as proof that 117 contracts are ready for your agent to call. Don't trust the Snap Store base URL on slices that belong to something else. And know what would move the number: one canonical OpenAPI per product with lifecycle markers, a reachable sandbox or public host for at least the Snap Store and Charmhub contracts, which already run as public services, and a well-known API catalog tying the 13 products together. The contracts are behind the documentation, and now you know where.
Read the original: https://apis.io/2026/10/27/canonical-holds-117-contracts-and-twelve-percent-of-them-are-callable/