FAVUREngine/ Agent usageread-only

Agent usage

snapshot of 2026-09-15 · Every agent role that ran, and how much work it did. · read-only
This is a frozen snapshot, not a live page. It shows one day and does not update. The live version runs on the operator's own machine and is not public. Figures are read from the engine's own stores for 2026-09-15; the snapshot was taken on 2026-09-15.
18agent roles activedistinct roles that worked
2,005agent instancesindividual agents spun up
19,992model requestsacross every role

Every role that ran

rolerequestsinstancesconversationsmodelsinput tokensoutput tokensreasoning tokenscache reuse
CODE5,69020604443,302,6422,174,765364,74471.1%
CODE_REVIEW3,4159504316,370,8733,070,9281,185,81675.4%
HEALTH_CHECK2,7701,3931,393348,489,425598,821122,88152.9%
DEVELOP1,4724903119,299,630623,841187,23065.2%
CRITIC1,282520391,826,6331,728,9711,010,60836.6%
PSEUDOCODE1,154480381,765,138769,192172,58675.1%
SCOUT7976262334,643,179386,02560,79482.3%
ORCHESTRATOR711120391,364,916249,43934,22173.2%
SPEC507200339,141,129543,24160,98473.8%
SPRINT_REVIEW48890346,869,122297,139133,93873.5%
CONCIERGE4801212219,797,341208,92917,04415.5%
SPRINT_PLAN33690335,735,521270,88460,79284.4%
ARCHITECT30270326,141,414174,16324,92064.4%
PLATFORM21460211,761,21053,8646,92282.6%
DIRECTOR1196037,857,35759,68012,76883.3%
BUILD9410034,435,02618,1021,94847.3%
TEST849032,758,19721,0222,48665.1%
UNKNOWN7700275,55466,59000.0%

One row per agent role. Reasoning tokens are the model's own deliberation, counted apart from what it actually wrote.

Day by day

dayrolesinstancesrequests
2026-09-15182,00519,992
2026-09-14186,30147,197
2026-09-13184,80138,084
2026-09-12186,21745,351
2026-09-11179,93573,140
2026-09-10177,71057,021
2026-09-09177,58755,837
2026-09-08174,81235,546
2026-09-07165,35937,678
2026-09-06165,87640,702
2026-09-05157,30256,168
2026-09-04132,19213,656
2026-09-03154,19932,046
2026-09-02153,25223,328
Counts, rates and ratios only. This page reports what the work was, never what it was billed at.