[strategist] v0.4 scope-substitution risk: i18n ✅ but llm-d/Drasi/kagent still at zero PRs — T-9 days

[clubanderson]

Strategic Finding

Type: priority-shift
Horizon: near-term (urgent — T-9 days)

The scanner is now executing v0.4 scope items — which is positive — but is prioritizing UX/accessibility work (i18n completeness) while the three Tier 1 core features remain at zero PRs:

v0.4 Scope Item	Tier	Status
llm-d stack monitoring (#18031)	1 — Core	❌ Zero PRs
Drasi reactive pipelines (#18032)	1 — Core	❌ Zero PRs
kagent/kagenti integration (#18033)	1 — Core	❌ Zero PRs
Nightly E2E expansion (#18034)	2 — Quality	❌ Zero PRs
Marketplace v2 (#18035)	2 — Quality	❌ Zero PRs
i18n completeness (#18036)	3 — UX	✅ First PR merged Jun 20
Accessibility audit (#18037)	3 — UX	❌ Zero PRs

Why This Is a Strategic Risk

PR #19321 merged June 20 closes #18036 — a legitimate v0.4 item. However, i18n completeness is a UX polish task that can ship without llm-d/Drasi/kagent. If the hive continues executing Tier 3 items (i18n, component consistency, accessibility) while ignoring Tier 1, v0.4 ships as a UX update rather than "AI-Native Observability." The milestone name becomes misleading.

The 1/7 progress number is a false positive: the one completed item is the lowest-impact item in the scope list.

Scope-Substitution Dynamic

The scanner naturally picks items where it can make progress independently:

• ✅ i18n — straightforward string wrapping, no external dependencies
• ❌ llm-d monitoring — requires llm-d cluster access, EPP routing understanding, real data hooks
• ❌ Drasi integration — requires Drasi CRD knowledge, change-feed architecture
• ❌ kagent/MCP — requires agent protocol expertise

Without a human DRI actively steering the hive toward Tier 1 work, automation will continue completing Tier 3 items while the milestone goal remains unmet.

Rationale

T-9 days remain. If the next 3-4 passes continue executing on UX items rather than feature work, the window for any meaningful Tier 1 delivery closes entirely. CNCF reviewers who evaluate v0.4 against the "AI-Native Observability" milestone will find only UX improvements — not the llm-d/Drasi/kagent integration that defines the milestone.

Proposed Next Step

1. Lock Tier 1 priorities: @clubanderson should comment on llm-d stack monitoring (EPP, model endpoint health, autoscaler dashboard) #18031, Drasi reactive pipelines dashboard #18032, kagent/kagenti full lifecycle management UI #18033 with explicit "start now" instruction
2. Pause Tier 3 scanner work: Redirect scanner capacity from i18n/accessibility toward creating stub implementations, card hooks, and test harnesses for llm-d/Drasi/kagent
3. Or explicitly declare scope-reduce: Acknowledge that v0.4 will not include Tier 1 features and update the ROADMAP milestone description accordingly

---

Filed by strategist agent (ACMM L6 — full mode)

[github-actions]

🐝 Hi @clubanderson! I'm kubestellar-hive[bot], an automation bot for this repo.

Trusted users — org members and contributors with write access — can mention @kubestellar-hive in a comment to trigger repo automation.
On issues, that mention queues an automated fix attempt. On pull requests, it records extra context for existing automation.
This is not an interactive Q&A bot, so mentions should be treated as requests for automation rather than a conversation.

Automation may take a moment to start, and follow-up happens through workflow activity rather than chat replies.