docs(planning): frontend milestones M6-M10 + alignment with M1-M5 - #1
Merged
Conversation
Introduce the frontend planning counterpart to M1-M5: - M6 Frontend Foundation (J3): design system, app shell, WS client, chat shell - M7 Control Room (J4): P&ID SVG, KPI bar, anomaly banner, wire chat, artifact registry - M8 Agentic Workspace (J5): 7 generative UI artifacts, Activity Feed, Agent Inspector with extended-thinking stream, onboarding wizard - M9 Polish & E2E (J6): work orders console, motion pass, memory scene, rehearsal - M10 Submission (J7): README, 3-minute video, submission package Also add ALIGNMENT.md which documents: - WebSocket event contract shared between backend and frontend - 4 new backend issues to add under zestones' milestones to secure prize eligibility (extended thinking on Investigator, agent-as-tool pattern, Q&A via Claude Managed Agents, 9 generative UI render_* tools) - Cross-team checklist to validate before starting J3
…pdate planning issues for M2, M3, M4, and M5
Move the 4 critical issues from ALIGNMENT.md into the actual backend milestones so @zestones sees them while working on M2/M4/M5 rather than only in a cross-reference document: - M2.9 — Declare 9 generative UI render_* tools in agents (enables inline artifacts in chat: SignalChart, DiagnosticCard, etc.) - M4.5 — Enable extended thinking on Investigator (Opus 4.7 flagship, streamed live via thinking_delta → shown in Agent Inspector UI) - M4.6 — Agent-as-tool pattern for dynamic handoffs (replaces asyncio-scripted pipeline with real agent-decided delegation: Investigator → KB Builder, Q&A → Investigator) - M4.7 (bonus P1) — Memory flex scene (Investigator uses failure_history to diagnose repeat patterns faster) - M5.4 — Migrate Q&A to Claude Managed Agents (with M5.2 as fallback) ALIGNMENT.md now points to the canonical location of each issue and keeps only the summary / rationale section.
…stones Backend (M4, M5): - M4.2 Sentinel: emit render_alert_banner ui_render event on anomaly - M4.3 Investigator: wire UI_TOOLS + ASK_KB_BUILDER_TOOL in tools_schema, dispatch render_*/ask_* in agent loop - M3.5: ask_kb_builder mini-session handler (enables M4.6 dynamic handoffs) - M3.6: render_kb_progress + render_equipment_kb_card UI tools for onboarding - M5.1 WO Gen: add UI_TOOLS + render_work_order_card integration - M5.2 Q&A: wire UI_TOOLS + ASK_INVESTIGATOR_TOOL, dispatch render_* and agent_handoff in agent loop (serves as fallback impl for M5.4) Frontend (M6, M9): - M6.4: add full EventBusMap (10 event types) and ChatMap (7 union variants) matching ALIGNMENT WS contract exactly - M9.1: WO console reacts to work_order_ready (live table refresh) and rca_ready (badge on row) via WS /api/v1/events
zestones
approved these changes
Apr 22, 2026
This was referenced Apr 22, 2026
vgtray
added a commit
that referenced
this pull request
Apr 24, 2026
Rewrite SCRIPT.md (v3 -> v4) and PRD.md (v2 -> v3) to follow zestones's
demo-plant-design.md on branch 133-m95-full-frontend-polish:
- SCRIPT v4: bottled-water plant, 5 machines (Source Pump, UV Sterilizer,
Bottle Filler hero, Bottle Capper memory, Bottle Labeler onboarding).
Grandparent Test applied throughout ("Motor shake", "Water pressure").
Scene 5 switches from Q&A to Memory recall on Capper.
Act III screen-records the live Agent Constellation.
Prononces "Opus 4.7", "Extended thinking", "Managed Agents", "MCP",
"17 tools", "1M context" in the voice-over.
- PRD v3: 6 demo endpoints (reset/full, reset/light, seed-forecast,
trigger-breach, trigger-memory-scene, run-full), DemoControlStrip
6 buttons, migration 010 rename plan, seed 7d history.
Backlog aligned on win-plan-48h Tier 0+1 (README rewrite #1 priority).
Hyperframes bonus section updated with 6-scene recordings structure.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Frontend planning counterpart to @zestones' backend milestones (M1–M5 pushed earlier today in 2c93a9b).
New milestones added
ALIGNMENT.md— must-read for both devsContains:
thinking_delta,agent_handoff,ui_render.render_*tools in agents — enables inline artifacts in chat (SignalChart, DiagnosticCard, WorkOrderCard, etc.)failure_historyto diagnose faster on repeat patterns)backend/*, frontend ownsfrontend/*)Why these 4 backend additions
The backend plan as pushed is technically solid (clean scoping, `DÉCIDÉ` markers, dependencies traced). But without these 4 additions, the multi-agent system would be an asyncio-scripted pipeline with no visible Opus 4.7 differentiator — losing the 2 most eligible $5k prizes.
Test plan
docs/planning/ALIGNMENT.mdand confirms the 4 new backend issues can be added under M2 / M4 / M5Closes no issue (planning-only, no code).