◎ Discry Score
okta.com
auth · API
B
0 / 100
DISCOVERY0
COMPREHENSION0
Category leader: 93 (A)
Discry your API →
AUTH · RANK #13 OF 16

Okta is Good to agents.

Discry independently scored how well an AI agent can discover and understand the Okta API from what’s public — not whether it’s usable. Below: every signal we checked, what’s costing the score, and what to change.

Discry your API — freeView the docs ↗

SCORED UNDER RUBRIC 1.2 · A full re-launch under Discry Score 2.5 — a new behavioral instrument, not comparable to these scores — is in progress.

Discovery

45% of score · 67/100
OpenAPI specA machine-readable OpenAPI/Swagger spec agents can parse.Pass
llms.txtAn llms.txt index that points agents to the docs that matter.Fail
llms.txt qualityThe llms.txt is focused, current, and well under the size budget.Fail
llms-full.txtA full-text bundle agents can load in one request.Fail
AGENTS.mdAn AGENTS.md that tells coding agents how to build on the API.Pass
.well-known/mcp.jsonA discoverable MCP manifest at a well-known path.Fail
MCP registryThe API is listed in a public MCP registry.Pass
robots.txt AI directivesrobots.txt allows (or explicitly guides) AI crawlers.Pass
SitemapA sitemap so agents can enumerate the docs surface.Pass

Comprehension

55% of score · 85/100
Task-oriented descriptionsEndpoints described by what they accomplish, not just their shape.Pass
Realistic examplesRunnable, real-world request/response examples.Pass
Multi-step workflowsDocs that chain calls into complete jobs an agent can follow.Pass
Error-recovery guidanceDocumented failure modes and how to recover from them.Partial
Answer-first formatThe answer leads; preamble does not bury it.Pass
Capability boundariesClear limits — what the API can and cannot do.Pass
Naming consistencyConsistent, predictable naming across endpoints.Pass
Heading hierarchyClean heading structure agents can navigate.Pass
Markdown docsDocs available as clean markdown, not JS-rendered HTML only.Partial
Token efficiencyDocs are concise enough to fit an agent context window.Partial

What we found

  • An agent can fully discover Okta's surface area: the complete Management API is published as a public OpenAPI spec (github.com/okta/okta-management-openapi-spec + the @okta/openapi npm package) and rendered in a modern reference portal, so capabilities are machine-readable rather than scraped.
  • Endpoint documentation is task-oriented and agent-friendly — descriptions lead with what you accomplish ("Creates a new user with a password and recovery question and answer. The new user is able to sign in after activation") rather than just the HTTP signature, and examples use realistic, consistent values (isaac.brock@example.com) instead of placeholders.
  • There is no llms.txt, llms-full.txt, or .well-known/mcp.json at developer.okta.com, so an agent gets no curated, token-efficient entry point — it must crawl the full portal or load the OpenAPI spec to orient itself.
  • Okta ships an official, first-party MCP server (announced 2025-09-22) and is listed in MCP registries (Glama), so an agent already has a sanctioned execution path beyond the raw REST API — a strong signal most APIs lack.
  • Capability boundaries are well-mapped: rate limits, 429/Retry-After behavior, compatibility and versioning rules, and deprecation policy are all documented, so an agent can reason about limits before hitting them rather than discovering them by failing.

What to change

Prioritized by impact on discoverability. You (or your docs platform) deploy these — Discry never touches your API.

  1. 01Publish an llms.txt at developer.okta.com that links the Core Okta API overview, authentication guide, OpenAPI reference, rate-limit, and error-code pages — the single highest-impact discovery fix, since the curated content already exists and only needs an agent-facing index.
  2. 02Add an llms-full.txt (or a markdown-content-negotiation path) covering core user/group/app/session flows so agents can ingest the essentials in <5K tokens instead of parsing the JS-rendered reference portal.
  3. 03Expand error documentation from descriptions to recovery steps for the top 10–15 error codes (e.g., which parameter failed on E0000001, exact remediation on 429) so agents can self-correct instead of just classifying the failure.
  4. 04Surface the official OpenAPI spec and MCP server directly from developer.okta.com discovery surfaces (a .well-known/mcp.json pointing at the MCP server, plus a prominent spec-download link) so agents find the sanctioned paths without a web search.
  5. 05Promote an AGENTS.md into the primary API/SDK repositories (it currently lives in okta/odyssey, the design-system repo) so coding agents working against the Okta SDKs get first-party guidance.

Execution coverage · INFORMATIONAL, UNSCORED

Whether an agent can actually complete a call and recover from errors is the deeper Audit layer — documented here, but not part of the Discry Score.

OAuth 2.0 (scoped access tokens, recommended)API token (SSWS)Private Key JWT Error format documented Rate limits documented Pagination documented Idempotency documented

Execution characteristics are documented thoroughly. Errors return a consistent machine-parseable JSON envelope (errorCode, errorSummary, errorId, errorLink, errorCauses) cataloged across ~250 codes. Rate limits are documented per-bucket with explicit 429 behavior and per-user thresholds; cursor-based pagination uses Link headers. General idempotency-key support is not documented.

See your own Discry Score.

Drop your API docs URL. See what an agent sees — in 60 seconds, free.

Discry your API — free