We took the vendors from 16 Gartner Magic Quadrants, covering API management, iPaaS, observability, cloud databases, security, coding agents and the rest, and profiled every one of them in the catalog. The result is the new Gartner section: 159 providers, grouped by quadrant and by Kin Score band. Gartner places a company as a vendor. We measure what it actually publishes for an integrator or an agent. Then we asked one question of all 158 that are scored: who is the most programmable?
Programmability, here, is two scores read together. The Kin Score composite measures the contract and everything around it: discoverability, contract quality and governance, operational transparency, developer ergonomics and access clarity. Agent Readiness measures whether an agent can actually drive the API, safely. The two run on different scales, so we rank each vendor by its percentile within the cohort on both and average the two. Neither score dominates, and a vendor cannot buy its way to the top on one alone.
The top ten
| # | Provider | Gartner placement | Kin Score | Agent Readiness |
|---|---|---|---|---|
| 1 | Harness | DevSecOps: Leader | 90.7 | 64.0 agent-native |
| 2 | CData | Data Integration: Niche Player | 84.4 | 66.5 agent-ready |
| 3 | Tray.ai | iPaaS: Visionary | 89.3 | 62.9 agent-ready |
| 4 | Cloudflare | Security Service Edge: Visionary | 80.6 | 72.5 agent-native |
| 5 | Infobip | CPaaS: Leader | 82.4 | 63.0 agent-ready |
| 6 | Telnyx | CPaaS: Niche Player | 79.6 | 83.6 agent-native |
| 7 | OpenAI | AI Coding Agents: Leader | 84.8 | 57.1 agent-native |
| 8 | Salesforce | Low-Code: Leader | 91.0 | 51.8 agent-ready |
| 9 | GitHub | AI Coding Agents: Leader | 80.3 | 59.9 agent-ready |
| 10 | Postman | API Management: Visionary | 79.6 | 54.9 agent-ready |
Harness wins on both axes at once. It is the only vendor in the cohort above 90 on the composite that is also agent-native, one of seven to reach that band. Salesforce has the highest composite of all 158, at 91.0, and lands eighth because its agent surface trails its contracts. Telnyx goes the other way: the highest Agent Readiness in the cohort, at 83.6, on a 79.6 composite. Half of the top ten are not Gartner Leaders. CData is a Niche Player, Tray.ai and Cloudflare are Visionaries, Telnyx is a Niche Player, and Postman is a Visionary.
Where Gartner and the APIs disagree
Seven of the 16 quadrants are won by a vendor Gartner did not name a Leader in that quadrant. Postman beats every API Management Leader, the best of which is Kong at #22. In Data Integration, CData at #2 sits above every Leader, and the best of those is Oracle at #23. Tray.ai takes iPaaS, where the best Leader, Workato, is #46. Cloudflare takes Security Service Edge ahead of Palo Alto Networks at #37. GitHub, a Challenger in Application Security Testing, wins it. Oracle takes Container Management as a Challenger. Snowflake, a Visionary in AI Platforms for Data Science and ML, beats Databricks, the best Leader there, at #31.
Placement still predicts something. Vendors that are a Leader in at least one quadrant have a median Kin Score of 66.0 and Agent Readiness of 36.5, against 56.1 and 29.5 for everyone else. The quadrants where the Leader does win are won clearly: Infobip in CPaaS, MongoDB in cloud databases, Dynatrace in observability, Okta in access management, Delinea in privileged access, plus OpenAI, Salesforce and Harness.
The bottom of the list holds Leaders too. Ab Initio, a Data Integration Leader, ranks #155 of 158. Proximus Global, a CPaaS Leader, ranks #157. Both are human-only on Agent Readiness: we found no public contract for either. Gartner rates what a company sells and how it executes. Neither of those requires an API anyone outside the account team can see.
The platforms that show up everywhere
| Provider | Quadrants | Rank | Kin Score | Agent Readiness |
|---|---|---|---|---|
| Oracle | 6 | #23 | 66.0 | 60.6 agent-ready |
| 8 | #34 | 70.1 | 44.1 agent-ready | |
| IBM | 8 | #49 | 69.1 | 36.3 agent-ready |
| SAP | 5 | #55 | 55.4 | 50.4 agent-ready |
| Microsoft | 10 | #73 | 69.2 | 27.1 agent-aware |
| Amazon Web Services | 9 | #87 | 59.6 | 30.5 agent-ready |
| Alibaba Cloud | 8 | #148 | 19.8 | 3.5 human-only |
| Huawei Cloud | 4 | #150 | 18.1 | 0.0 human-only |
Breadth is not programmability. Microsoft appears in ten quadrants and AWS in nine, and both land in the middle of the cohort. Oracle is the most programmable of the hyperscale vendors on this reading, mostly on agent readiness. The two Chinese clouds sit in the bottom dozen, despite Leader placements between them in cloud platforms, cloud databases and container management.
A note on fairness. The Kin Score rates a company, not a product line, so Microsoft carries one score into all ten of its quadrants. The one exception is API Management, where Azure API Management has a listing of its own. Agent Readiness also does not yet credit several first-party MCP servers that the catalog lists but has not yet verified, including those from AWS, Microsoft, SAP, Oracle, Databricks and Kong. Those six will move up when it does, and the Gartner page will move with them.