LearnNewsExamplesServices
Frontmatter
id14599
titleFM cockpit fleet grid + health summary bar (density-informed layout)
stateClosed
labels
enhancementdeveloper-experienceai
assigneesneo-opus-vega
createdAtJul 4, 2026, 4:16 AM
updatedAt3:16 AM
githubUrlhttps://github.com/neomjs/neo/issues/14599
authorneo-opus-vega
commentsCount2
parentIssue14560
subIssues[]
subIssuesCompleted0
subIssuesTotal0
contentTrust
projected
quarantined0
signals[]
blockedBy[ ] 14598 FM cockpit AgentCard component — the resident card at the artifact bar
blocking[]
closedAt3:16 AM

FM cockpit fleet grid + health summary bar (density-informed layout)

Closed Backlog/active-chunk-3 enhancementdeveloper-experienceai
neo-opus-vega
neo-opus-vega commented on Jul 4, 2026, 4:16 AM

Context

T2.8 of Epic #14560 (map v1.1) — the cockpit's default view: the fleet grid of AgentCards with the health summary bar, laid out against the measured density evidence (#14592) rather than the mock's assumptions. Filed under the decomposition sprint (D#14561).

The Problem

The SSOT mock renders 6 agents in 2 columns with a 3-count health strip. Measured reality (#14592, 2026-07-04): 12 registered identities, 6 peak-concurrent, category counts swinging 1→6 within 2 hours, card lane-lines churning every 10–25 minutes at sprint pace. A mock-faithful layout breaks at live scale — this leaf is where the density data becomes structure.

The Architectural Reality

  • Composes T2.7 AgentCards; consumes #14571 roster DTO + #14595 runtime-status for the counts; namespace per #14577.
  • Density-derived requirements (#14592 evidence, sourced there): 3-column desktop breakpoint OR ranked-fold (online-first, idle collapsed) at 12+ registered; health bar animates count transitions (state churn is the norm); benched/guest render within the 5-state legend (no 6th category).
  • The health bar is the scale-to-a-glance instrument (#13444 carry-forward): counts must stay legible before any card is read.
  • Grid rendering rides engine strengths — this is a live-data grid the engine's update rate exists for; no virtual-scroll library imports (sibling-pattern: engine-native).

The Fix

One PR: (1) FleetGrid (responsive columns + ranked-fold behavior, config-driven thresholds citing #14592) + HealthBar (5 categories, animated count transitions, reduced-motion honored); (2) unit specs — fold ordering (online>idle>benched), threshold behavior at 6/12/20 agents (fixture rosters), count-transition config; (3) NL-verifiable mount bound to the live roster.

Acceptance Criteria

  • Grid lays out 6 / 12 / 20-agent fixture rosters per the density-derived rules (3-col or ranked-fold at 12+; thresholds cite #14592).
  • Health bar renders 5 categories from #14593 swatches; count changes animate (reduced-motion honored).
  • Fold ordering: online first, idle collapsed group, benched/guest tail — deterministic.
  • Live wire binding for counts (#14595); stale-as-stale degrade.
  • Unit specs (custom playwright config) + NL-verifiable render at live roster scale.

Out of Scope

AgentCard internals (T2.7) · activity stream (T3) · shell/perspective composition (T6, #13158 contracts) · wire methods (Lane 2).

Related

Epic #14560 (parent) · blocked-by T2.7 (filed alongside) + #14593 · consumes #14571/#14595 · #14592 (the evidence this consumes — every threshold cites it) · #13444 (scale-to-a-glance) · D#14561.

Live latest-open sweep: latest 8 checked 2026-07-04T02:13Z (newest #14595); no equivalent. A2A sweep: clean. Structure-map gate: N/A — app-layer view.

Origin Session ID: a28f1415-37a4-4a5a-a3e4-b56adf6274ec Retrieval Hint: "FM cockpit fleet grid health bar ranked fold density thresholds"

tobiu referenced in commit 1bc0708 - "feat(agentos): fleet cockpit StateDot session-state primitive (#14593) (#14700) on Jul 4, 2026, 9:13 AM
tobiu referenced in commit 5b6b65d - "test(agentos): fleet grid at density-evidence scale — NL store possession proves rank/fold/counts on the mounted surface (#14599) (#15023) on 3:16 AM
tobiu closed this issue on 3:16 AM