Mercury · Payments◎ discry
Legible
Usable, with real gaps
DISCOVERY60.5
COMPREHENSION75.6
Small-model ready#15 of 34 Payments APIs · 59th percentile
PAYMENTS · RANK #15 OF 34

Mercury is Legible to agents.

Discry independently scored how well an AI agent can discover and understand the Mercury API from what’s public — measuring discovery and comprehension separately, not a single blended number.

Familiarity 67% · what a model knew before reading the docs

Discovery

61 / 100
OpenAPI specfound at https://mercury.com/api-docs but invalid: unparseablePartial
llms.txt102 described linksPass
llms-full.txtnot presentFail
AGENTS.mdnot found in repoFail
.well-known/mcp.jsonnot presentFail
MCP registrysmithery match for "mercury"Pass
robots.txt AI directivesparsed AI-bot rulesPass
Sitemapincludes docs pagesPass

Comprehension

76 / 100
Factual recall0/2 passed
Request construction1/1 passed
Absence honesty1/1 passed

Where agents fail

What HTTP header (exact header name) must carry the API credential on requests to this API? 3/3 runs failed

How does a client target this API's test or sandbox environment (e.g. a distinct key prefix, separate base URL, or header)? 3/3 runs failed

See your own Discry result.

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

Discry your API — free