docs(reshape): local-first dual-mode plan + reconciled research + Notion doc

[gnanirahulnutakki]

What this is

The consolidated, reshaped Sith plan: it reconciles the two research lanes into one
authoritative synthesis and reshapes the plan docs + epics to the research-backed local-first
dual-mode direction, plus a self-contained Notion-ready page.

How to review: the branch has two commits. a550c4b assembles the planning baseline (the
existing plan docs + M0 + epics + research, imported unchanged from their sibling branches so the
reshape shows as clean diffs). 6b81428 is the reshape — that is the commit to review.

Reconcile decision (the RECONCILE step)

docs/research/USE-CASE-AND-SHAPE.md is now one authoritative document that keeps PR #16's
evidence and adopts PR #15's sharper day-0 framing (local-first single-binary CLI/TUI as
the day-0 wedge; MCP server in v1, not Phase 3; honest sizing of the read wedge; the k9s/Aptakube/
Replit/DORA/Backstage-TCO evidence).

➡ Recommend closing PR #15 (docs/research-usecase) — its synthesis is fully absorbed here.
PR #16's other three evidence files (market, identity, integrations) are unchanged and carried on
this branch, so #16 can also be closed in favour of this consolidated reshape (or merged
first, then this rebased). See "Relationship to the other PRs" below.

The reshaped direction, in brief

One Go binary, three modes: sith (CLI/TUI — the k9s-style day-0 wow) · sith ui (local web
"fleet IDE", kubeconfig-direct, no account/telemetry — "Lens but better") · sith hub (day-N
federated: OCM minions, Workspace isolation, Ardur PDP, governed MCP write surface). One
source-abstract fleet model (local kubeconfig or OCM spoke) and one enforcement pipeline.

Roadmap: Phase L (day-0 local mode, no OCM, ships first — the adoption wedge, plus MCP
read tools) → M0 (done) → P1 hub read federation → P2 first governed write (gitops.open-pr + its
MCP write tool) → P3 policy federation + full MCP write surface. Fast-follow: connector framework
(E12), cost overlay (E13), multi-cloud/air-gap. The moat is governed action + governed agent
access; the local read view is the on-ramp — you earn the right to govern a fleet by first being
the tool the engineer already uses to see it.

What changed

File	Change
docs/research/USE-CASE-AND-SHAPE.md	Reconciled synthesis (supersedes both prior versions)
docs/CHARTER.md	Two wedges; individual operator as top-of-funnel; adoption success criterion; local-first thesis claim
docs/SCOPE.md	Fixes the Lens contradiction — OUT "another single-cluster console", IN "aggregated multi-cluster local client"; contractual non-goals (no re-skinning / metering engine / agent-orchestration framework)
docs/ARCHITECTURE.md	Three run-modes; source-abstract read federation; four-mode identity; connector framework; cost overlay; cosign/SLSA/SBOM; SPIFFE-without-SPIRE
docs/ROADMAP.md	Phase L (day-0 local, no OCM) ahead of the hub track; MCP read in Phase L
docs/EPICS.md	New E11 (local fleet client), E12 (connector framework), E13 (cost overlay), each with a flowchart per feature; E2 source-abstract; E7/E8/E9 reshape notes; updated roadmap map + open questions
docs/SITH-NOTION.md	New — one self-contained Notion-ready page: overview + wedge + form factor + prioritized roadmap + carry/discard/net-new + every epic (E0–E13) with per-feature flowcharts + open questions

Relationship to the other PRs (for the owner to decide)

This branch is a superset consolidation — it carries reshaped copies of the initial plan
(#1), the epics (#14), and the research (#15/#16). Two clean paths:

• (A, simplest) Treat this as the canonical consolidated plan: merge this, close docs: initial Sith plan — charter, architecture, ADRs, threat model, roadmap, competitive #1, docs: implementation-epics plan (EPICS.md) #14,
  docs(research): use-case & shape — what Sith should be to become a default tool #15, docs(research): full research set — use-case & shape, market/form-factor, identity, integrations #16.
• (B) Merge docs: initial Sith plan — charter, architecture, ADRs, threat model, roadmap, competitive #1/docs: implementation-epics plan (EPICS.md) #14/docs(research): full research set — use-case & shape, market/form-factor, identity, integrations #16 first, then rebase this branch so its diff shows only the reshape;
  close docs(research): use-case & shape — what Sith should be to become a default tool #15 regardless.

Either way, #15 is superseded.

Needs the owner's decision (new open questions, EPICS §9)

• Local-mode hero (Q12): TUI-first, sith ui-first, or both together?
• Local→hub upgrade UX (Q13): re-import, side-by-side, or in-place promote when a
  kubeconfig-direct cluster becomes an OCM minion?
• MCP-read auth in local mode (Q14): loopback trust, local token, or keychain secret?
• Telemetry stance (Q15): permanent hard-no, or a disclosed off-by-default opt-in later?

No product code; docs only. All relative links resolve on-branch.