Files
stack/docs/fleet/proposals/mosaic-platform-prd/PRD-jarvis-main-agent.md
Jason Woltje 6b8c3c5d3a
All checks were successful
ci/woodpecker/push/ci Pipeline was successful
docs(fleet): refine Mosaic Platform PRD per Jason's review (R1–R7)
Post-review refinements to the DRAFT PRD set (decisions D1–D12 unchanged):

- R1: PRD-permission-relay is now self-contained — its cited design source
  (guard-rails-capability-permissions.md) is absent from origin/main, so the
  essential model (prepare-freely/execute-with-approval, permission levels,
  resource:action grants) is folded in as the authoritative spec.
- R2: Phase-1 conversation channel is tmux/CLI via the f4 Phase-1 tmux
  connector (operator issues /remote-control); Matrix room (J5/K1) is Phase 2.
- R3: Jarvis is a Level-0 orchestrator (delegation + subagents, never executes
  coding/infra); AC-NS-8 latency isolation is a separate-model-capacity
  guarantee, not just a separate process (NS-10, AC-NS-8, ASM-5).
- R4: jarvis-brain reframed as the P0 (prototype) Mosaic Stack; tenant-1
  migration targets the proper stack storage layer (vector DB + enhanced
  memory + flat-file backends). Agent memory/runbooks migrate live into the
  memory subsystem and are NOT frozen; repo retires read-only only after both
  PA data and agent memory are verified migrated (X2/X-R7).
- R5: homelab→USC promotion is an explicit owner sign-off gate, not an
  automated metric (X deployment-scope).
- R6: verified K1 is covered by f4-matrix-connector.md Phase 2 (no new doc).
- R7: clarified always-available Opus ≠ standing cost (idle = no bill) (J-R2).
- Also normalized PRD-backlog-providers.md and PRD-webui-fleet-control.md with
  prettier (whitespace-only) — they were non-conformant as originally pushed
  and would have re-broken repo-wide format:check on merge.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RMoEx7hfdFGjUiCHuN1RRi
2026-07-09 14:50:54 -05:00

10 KiB
Raw Blame History

PRD — HMI Main Agent ("Jarvis") · Workstream J

Status: DRAFT for ratification · Source of truth once landed: NORTH_STAR.yaml goals J1J5 Depends on upstream: H2 (system-type profiles), A3a (card lifecycle), B1 (supervisor tick), F4/K1 (Matrix connector)

Mission

Every Mosaic system gets exactly one always-on human-machine-interface agent — default alias Jarvis, unit mosaic-agent@main.service — that owns the human relationship: conversation, idea development, schedule, email, tasks, knowledge. It delegates all engineering/research/ops work to the orchestrator (Mos, mosaic-agent@orchestrator.service) through the Mosaic Backlog, and reports fleet status to the user without ever interrupting the orchestrator.

Jarvis is a Level-0 orchestrator: it accomplishes its own work through delegation and subagents, never by executing coding/infra tasks itself. PA mutations (tasks/events/knowledge) are direct API calls; everything heavier is either a spawned subagent (research, drafting, analysis) or a backlog card handed to Mos (engineering/infra/fleet). This keeps the main agent's context conversational and light.

This solves the observed failure mode: a busy orchestrator that can't respond, accumulates conversational context rot, and derails over time. Post-split, the orchestrator's context is execution-only.

Because Jarvis and Mos are separate agents with separate model capacity (D11: Jarvis on Opus, Mos on Fable; independent inference quota), orchestrator load cannot degrade conversational latency — the isolation in AC-NS-8 is a capacity guarantee, not merely a separate process.

Requirements

Persona & runtime (J1)

ID Requirement
J-R1 Jarvis is provisioned from the personal-assistant persona baseline via system-type profiles (H2/H3); alias, model tier, host, and channel are profile fields, not code.
J-R2 Default model tier Opus (ratified D11); the orchestrator's tier is independent. Always-available ≠ always-billed: Opus is provisioned 24/7 but cost is per-interaction — an idle Jarvis (no user turn in flight) incurs no model spend, so "always-on" carries no standing token bill.
J-R3 Jarvis survives reboot under systemd (mosaic-agent@main), participates in the fleet heartbeat protocol, and is counted in the supervisor's health floor.
J-R4 Persona customization is update-surviving per H4 (override layer wins on merge).

PA toolchain (J2)

ID Requirement
J-R5 Jarvis executes personal-assistant mutations directly in the user's workspace via the product API: tasks, events/calendar, knowledge entries, ideas, tickets. No delegation for PA ops (ratified D4).
J-R6 External PA integrations (email, external calendars, helpdesk) are workspace-scoped integrations with credentials in the product credential vault; actions flagged requires_approval route through the permission relay (workstream P).
J-R7 Until tenant-1 migration (X2) completes, Jarvis may read/write the jarvis-brain flat files as a transitional adapter; the adapter is deleted at X2 cutover.

Delegation contract (J3)

ID Requirement
J-R8 The Jarvis→Mos handoff is only via Mosaic Backlog cards: goal, acceptance criteria, priority, depends_on, advisory budget. Never via chat messages to the orchestrator.
J-R9 Jarvis translates conversation outcomes into card sets; ambiguity is resolved with the user before card creation — the orchestrator receives only decision-complete work.
J-R10 Card authorship is attributed (author=main-agent, ratified-by=user where applicable) for audit.
J-R11 Authority line: Mos holds all execution and merge authority (NS-4). Jarvis relays the user's GO/NO-GO gates as card state, and never acquires fleet mutation, merge, or dispatch rights.

Passive observability (J4)

ID Requirement
J-R12 Jarvis answers "what's the fleet doing" from read-only sources: heartbeat files, mosaic fleet ps JSON, backlog card states, CI status. Zero messages to the orchestrator for status.
J-R13 Jarvis proactively surfaces to the user: blocked cards, failed CI on user-ratified missions, approval requests pending, budget advisories. (PDA-friendly phrasing per SOUL.md.)

Channel (J5)

ID Requirement
J-R14 Phase 2 (target channel): Jarvis's conversation lives in a dedicated Matrix room on the self-hosted homeserver via OrchestratorConnector(matrix) (K1 = f4 Phase 2). Matrix-first: no Discord channel is created for Jarvis (ratified D1).
J-R14a Phase 1 (interim channel, ratified): Jarvis runs on the tmux/CLI connector — the f4 Phase-1 default connector. The operator launches the mosaic-agent@main tmux session and issues /remote-control to grant interactive access; this is the day-one conversation surface. No Discord, no Matrix dependency in Phase 1 (keeps D1 intact and unblocks J1J4 before K1 lands).
J-R15 Multi-platform user reach arrives via mautrix bridges (K2); Jarvis's code path is Matrix-only (from Phase 2 onward).

Acceptance criteria

  1. AC-NS-8: user converses with Jarvis under full orchestrator load; latency unaffected; orchestrator receives zero conversational traffic.
  2. AC-NS-9: a conversationally-agreed mission round-trips (cards → drained → completed → reported by Jarvis) with no chat handoff.
  3. Kill the orchestrator mid-conversation: Jarvis conversation is unaffected; Jarvis reports the outage from heartbeat state. (Directly exercises the separate-capacity guarantee.)
  4. !sys-equivalent admin verbs work in Jarvis's active channel — the tmux/CLI session in Phase 1, the Matrix room in Phase 2 (status/logs/clear/restart of the main agent).
  5. Phase-1 channel: operator launches the mosaic-agent@main tmux session, issues /remote-control, and holds a full conversation with Jarvis over CLI with no Matrix/Discord dependency.

Non-goals

  • Jarvis executing code/infra changes (that is Mos + fleet).
  • Horizontal sharding of the main agent (rejected in the Matrix charter: split-brain).
  • Per-workspace fleets (post-MVP per ASM-6).

Open items (for Mos's planner)

  • Context hygiene: Jarvis's durable memory is the workspace (tasks/knowledge/ideas); define its resume protocol (KICKSTART-equivalent) so /clear is cheap. ASSUMPTION: mirror the MOS-KICKSTART two-file pattern until the product grows a native session-resume feature.
  • Reconcile the old apps/api matrix-bot-sdk workspace bridge with the F4 connector design (one Matrix stack, not two). NOTE (verified 2026-07-09): no matrix dependency remains in apps/api on origin/main — this item is likely already moot; confirm before K1 build.