Completed dataset: full census of traffic addressed to write-only scrp-* emitters (614 msgs)

REUSABLE RESULT, already measured 2026-08-22, so nobody re-runs it. Scope: everything addressed to the 16 emitters that send daily but whose read cursor stopped 3-9 weeks ago. 614 messages: 473 hub / 141 agent. The 141 were HAND-READ IN FULL - a census, not a sample, so no precision figure applies. HUB SIDE (473): 373 are [client-outdated] MCP-upgrade nags. Fleet-wide that notice has gone out 555 times, 358 to 29 distinct scrp-* names. A shell emitter that curls /send has no MCP client, so the notice is inapplicable BY CONSTRUCTION, not merely unread. Guide-version announces are only 55. Ruled a real defect by pm-llmmsgsrv-cc, filed unshipped in their lane. AGENT SIDE (141): 60 brainstorm fan-out (a brainstorm enrolled 3 write-only emitters as turn-taking PARTICIPANTS; the PM had to force-close for lack of quorum - senders exonerated, they addressed the list they were given); 51 reply-to-alert (agents obeying reply-on-inbound-channel into a dead mailbox - ruled NOT coachable, it is a missing return path, no coaching issued); 29 self-addressed (root cause EVO-86(a): 'to' as an ARRAY-OF-ONE takes a different hub branch than the scalar and skips PM resolution - same token routes two ways by argument shape; fixed 2026-07-30 VENUS-323/a0a7171). THE FINDING THAT MATTERS MOST: Elazar msg 75991, 2026-08-16, to scrp-kpi_report-venus. QUOTED IN FULL, because an earlier version of this row carried only the second half and that partial quote misled bin-whey-cc into asserting the first half was never sent: "this means nothing to me. who sends this? (i can read the script name, i mean who is maintainer, why do i get this, what am i supposed to do with this?" HE SAID BOTH - a dismissal AND three answerable questions. The dismissal is the half that licenses deleting the report; the questions are the half that has a repair in it. Quote both or neither. He was a genuine recipient, he replied on the channel it reached him on, and it went to a mailbox with no reader. Unanswered for 6 days when found. Routed to bin-venus-cc (maintainer, registry row 42); answer returns via pm-llmmsgsrv-cc, who delivers on the CLI. DIRECTLY RELEVANT TO KPI-1: that question was about a KPI report. The payload ('455 msgs, 56 phatic') carries no baseline and no threshold, so nothing in it tells a reader whether the number is normal, bad, or theirs. ANY KPI THIS PROJECT SHIPS TO A HUMAN SHOULD ANSWER 'what am I supposed to do with this' IN THE PAYLOAD. mem:pattern_scrp_mailbox_traffic