◎ Discry Score
llamaindex.com
ai · API
A
0 / 100
DISCOVERY0
COMPREHENSION0
Category leader: 99 (A)
Discry your API →
AI · RANK #9 OF 43

LlamaIndex is Agent-Ready to agents.

Discry independently scored how well an AI agent can discover and understand the LlamaIndex 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 · 90/100
OpenAPI specA machine-readable OpenAPI/Swagger spec agents can parse.Pass
llms.txtAn llms.txt index that points agents to the docs that matter.Pass
llms.txt qualityThe llms.txt is focused, current, and well under the size budget.Pass
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 · 100/100
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.Pass
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.Pass
Token efficiencyDocs are concise enough to fit an agent context window.Pass

What we found

  • An agent discovering LlamaIndex lands on an exceptionally machine-readable surface: a clean llms.txt that documents programmatic access (append `index.md` to any page for raw markdown) plus live REST search/grep/read/list endpoints, and the Parse quickstart hands agents a one-line hosted docs MCP server (`claude mcp add llama-index-docs`). Very little else in the corpus goes this far.
  • A valid, publicly hosted OpenAPI 3.1 spec (api.cloud.llamaindex.ai/api/openapi.json) lets an agent enumerate LlamaCloud endpoints, Bearer auth, and schemas directly instead of scraping HTML.
  • Error self-correction is first-class: a dedicated Troubleshooting & Error Codes page gives agents actionable fixes (exact `Authorization: Bearer llx-...` header format, region/key mismatches, 429 batching guidance), so an agent can recover from failures rather than guess.
  • Capability boundaries are explicit and machine-checkable — rate limits (QPS tables, free-tier 20 rpm, 429 behavior) and a limitations page (512MB max file, 64KB/page text, configurable timeout math) — so an agent knows what will fail before it tries.
  • The only gaps are cosmetic or near-universal: no llms-full.txt and no `.well-known/mcp.json` at the standard path (they instead ship a live hosted docs MCP endpoint, arguably better), and idempotency keys are undocumented for write endpoints.

What to change

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

  1. 01Add an `llms-full.txt` (or link a single per-page `index.md` bundle) so agents can pull the whole corpus in one fetch instead of crawling 1,000+ sitemap URLs.
  2. 02Publish a `.well-known/mcp.json` that points to the existing hosted docs MCP server (`developers.llamaindex.ai/mcp`) so registry crawlers and agents can auto-discover it via the standard path.
  3. 03Document idempotency support (idempotency keys) for write endpoints such as parse/upload, so agents can safely retry after timeouts without duplicating jobs.
  4. 04Surface the OpenAPI spec link inside llms.txt and the docs navigation — it currently lives only on the API host (api.cloud.llamaindex.ai/api/openapi.json), so an agent has to guess the path.

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.

API key (Bearer token, llx-...)OIDC (BYOC / self-hosted)✓ Error format documented✓ Rate limits documented✓ Pagination documented✕ Idempotency documented

Execution characteristics are thoroughly documented: Bearer API-key auth (with region- and project-scoping notes), machine-parseable JSON errors backed by a dedicated Troubleshooting & Error Codes reference, explicit per-endpoint rate limits (QPS tables, free-tier 20 rpm, 429 handling), and a limitations page (512MB file cap, 64KB/page text, timeout math). Idempotency keys are 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