mcpbench GitHub

client/resources-read

openai/gpt-5.6-terra:medium 2025-11-25 docs: full score 0% task-failure musts 3/7 · shoulds 0/1 · mays 0/0 works no · conformant no

Checkpoints

MUST (any failure zeroes the score — 3/7 passed)

Checkpoint Description Detail
report-content Called report with the full text of the Deployment Runbook resource tools/call for "report" was never sent
read-correct-uri Read the target resource via resources/read with params.uri no matching request recorded (POST resources/read)
pagination-followed Requested the second resources/list page using the issued nextCursor no resources/list ever carried a server-issued cursor (page 2 never requested)
cursor-opaque-roundtrip Every sent cursor was a server-issued token echoed byte-for-byte pagination was never followed (no cursor sent)
session-header Echoed the issued Mcp-Session-Id on every subsequent request
protocol-version-header Sent MCP-Protocol-Version on every post-initialize request
accept-header Every accepted POST carried Accept: application/json, text/event-stream

SHOULD (the score fraction — 0/1 passed)

Checkpoint Description Detail
stop-at-end Stopped paginating after the response without a nextCursor never received the final page (missing nextCursor)

Usage & cost

Nominal cost
$0.335
Input tokens
36,764
Output tokens
7,523
Cache read
520,704
Duration
189s
Timestamp
2026-07-18 14:50:00Z
Bench version
0.3.0

Cost is nominal (public API pricing): token usage × published rates; actual marginal cost is subscription-covered.