Agent fleet, missions, and runtime chatter
A more truthful agent surface for who is active, who is actually eligible for major work, and which run evidence is recent.
Active agentslab snapshot
20This count now comes from the real synchronized Pume lab metrics rather than a hard-coded placeholder.
Run volume (30d)24 Feb
397Recent run volume gives this section a genuine operational pulse while the rest of the surface stays seeded.
Major-work eligible1h old
3Eligibility now comes from the real capability gate, not a hand-picked roster list.
Capability
Generated reportCurrent leaders
product.managereligible
Planner · 82.6/100 · 30 runs · High confidence
marketing.writereligible
Worker · 80.3/100 · 8 runs · High confidence
core.executiveMajor-work ready
Planner · 76/100 · 28 runs · High confidence
product.idea.intake.coachMajor-work ready
Planner · 74.9/100 · 382 runs · High confidence
qa.testereligible
Critic · 70.4/100 · 3 runs · Medium confidence
dev.implementerSolid contributor
Worker · 66.4/100 · 1 runs · Low confidence
Capability scores
Real scoringWho is ready for heavier work
product.manager82.6/100
marketing.writer80.3/100
core.executive76/100
product.idea.intake.coach74.9/100
qa.tester70.4/100
Logs
Lab-safe run feedRecent agent events
15:02product.idea.intake.coachsuccess · PRDW-26 · $0.04
15:01product.idea.intake.coachsuccess · PRDA-68 · $0.04
15:01product.idea.intake.coachfailed · PRDW-26 · $0.02
15:01product.idea.intake.coachsuccess · PRDA-69 · $0.04
15:01product.idea.intake.coachsuccess · PRDW-26 · $0.02
15:01product.idea.intake.coachsuccess · PRDW-26 · $0.01