[Release]: PLATE v1.0.0 Roadmap — Endless Q&A-driven autonomous product lifecycle (far-future target)

Dedicated Release/Roadmap for PLATE v1.0.0 (far-future target).

This is now the primary container for the v1.0.0 vision and readiness criteria (moved from #627 per planning session).

Q&A Round Status (2026-06-17): This interactive Q&A refinement round is paused per user direction. #654 now serves as the current summarized container for the full v1.0.0 vision, all captured refinements, sub-Epics (#656–#661), detailed stories, metrics, risks/non-goals, and supporting artifacts (e.g., design doc). No further immediate changes from this round. Future Q&A sessions can resume refinement of the vision.

Autonomous v1.0.0 Completion Plan (2026-06-18): Approved plan targets all ~98 open issues (23 Epics incl. this, 49 Features, etc.) for 1.0.0. See local plan.md (or session artifact) for full: phases (setup, core PM/Autonomy/Feed #660/#657/#656, loops, adoption, polish, ceremony), human checklist (budgets/reviews/merges), order, reuse (plate_* tools, autonomy.py, AGENTS loops), verification (health, #654 checklist, close all).

Group: ~58 v1.0 direct (refs to 654/660/657/etc.); fold legacy (payload-adoption:8, spec-audit:6 etc.) + bugs/docs. Use #654 as target Epic.

End-user experience vision

The system endlessly feeds the user Questions and Tasks. By engaging in Q&A and completing these tasks, the user is able to specify, develop, deploy, monitor, and market their software product. The details of this process are, as much as possible, opaque (but accessible) to the end user.

Release 1.0.0 readiness criteria (all must be true)

Truth-up (2026-07-24, Phase 0.2 of v1.0 execution plan): Checkboxes require production-complete behavior with E2E proof — not MCP/test sketches. Items previously marked [x] that only have stubs/sketches were demoted to [ ] with a Partial note. Do not re-check without verification against acceptance on the linked Feature/Epic.

Core workflows (entirely Q&A driven)

• There is a "product planning" workflow, driven entirely by Q&A — Partial: planning schema / MCP sketches; Feature #628 still status:stub (Epic #656)
• There is a "release planning" workflow, driven entirely by Q&A — #629, #640
• There is a "feature planning" workflow, driven entirely by Q&A — #630

User-facing surfaces

• There is an interface for surfacing arbitrary Questions to the user — Partial: run-qanda-session + some PM feed stubs; endless feed incomplete (#631)
• There is an interface for surfacing arbitrary Tasks to the user — Partial: PM stubs; Task features #359/#360 still need refinement (#631)
• The system can surface Design visualizations and documents for user approval — #632, #646
• The system can surface Research results for user approval — #632

Adoption & bootstrap

• The system can be easily integrated into an existing repository with minimal friction (under 30m of onboarding) — #633, #615+#616–#622
• The system can bootstrap brand new projects — #633 + bootstrap CLI (no timed proof metrics yet)

Budgeting & safety

• The user can budget resource usage for autonomous tasks, and the system safely operates within those budgets — Partial: AutonomyEngine + costs MCP (#470 family); feed enforcement incomplete (#634, #653, #645)

Evidence & demos for humans

• The system generates release notes with videos or animated GIFs — #635 (render tests sketch media fields)
• Every new Feature has explanatory videos or animated GIFs (generated from automated test runs) that the user approves — #636

Full autonomous capabilities (the system can autonomously...)

• Author a stubbed Release, Feature, Bug, Epic, Design, Research, Question or Task Issue based on Q&A input — Partial: plate_plan_epic and issue tooling exist; full type coverage + Q&A drive incomplete (#637)
• Refine a stubbed issue through further Q&A input — Partial: same (#637)
• Resolve a bug (open a PR draft, add failing tests, get tests passing, take the PR from draft to ready, work the review-and-respond cycle, and merge) — #638 (+ agent bugs #508–#512, #605, #496)
• Implement a feature (estimate cost, open a PR draft, add failing tests, add failing design validation tests, get tests passing, etc.) — #639, #646
• Plan an epic — Partial: risk-aware plate_plan_epic (#477 closed); not full Q&A-driven loop (#640)
• Plan a release — #629, #640
• Perform refactor and rearchitecture tasks on a scheduled frequency — #641, #659
• Deploy to production — #641 (sim/feed sketch only today)
• Implement an epic — #641, #660
• Cut and finalize a release — Partial: cut/finalize CLI automation in 0.7.x; not PM-orchestrated E2E (#641)
• Implement a release — #641
• Deploy a marketing site — #641
• Package software and deploy to a marketplace — #652, #384 + human Tasks #380/#381/#626
• Review discussions on a scheduled frequency, turning some into Issues — #642
• Monitor market conditions on a scheduled frequency, surfacing concerns to the user — #642

Collaboration

• The system can play nicely with human developers on GitHub — #643, #651

Additional must-haves (from refinement session)

• Multi-agent orchestration + delegation surfaces (#644) — Partial: delegation helpers/MCP sketches; Epic still stub
• Simulation / dry-run / shadow mode for high-impact actions (#645) — Partial: autonomy.py + tests sketch; not a hard gate on high-impact actions
• First-class design validation + visual/interaction contracts (#646)
• Provenance + decision ledger (#647)
• Consistent checkpoint / approval primitive (#648) — Partial: MCP stub + tests; not unified across all surfaces
• Self-updating / self-migrating PLATE (#649)
• Hybrid/non-code project support (#650)
• Stronger human+agent branch etiquette & signaling (#651)
• Release/marketplace packaging with first-class media + adoption proof (#652)
• Cost/risk observability dashboards driving the feed (#653) — Partial: stub + tests; not driving prioritization

Current sub-Epics (thematic groupings under #654)

Formal hierarchy repaired 2026-07-24 (Phase 0.1): GitHub sub-issue links now match the program tree (previously text-only for most children).

Direct children of #654 include: #656, #657, #658, #659, #660, #633, #637, #638, #639, #643, #650, #627 (pointer), #384 (marketplace human path), #390 (licensing).

• #656 Q&A-Driven Planning Workflows & Endless Feed Surfaces → children #628, #629, #630, #631, #632, #640
• #657 Autonomy Foundations → children #645, #648, #634, #653, #647 (multi-agent #644 lives under #660; design validation #646 under #639)
• #658 Evidence, Media, Packaging → children #635, #636, #652
• #659 Scheduled Procedures, Monitoring & Deploy → children #641, #642
• #660 Project Manager / Orchestrator (core v1.0.0 deliverable) → children #661, #644
• #633 Frictionless adoption → children #615, #649
• #638 Full autonomous bug resolution loop
• #639 Full autonomous feature implementation → child #646
• #637 Autonomous authoring/refinement of stubbed Issues
• #643 Human collab → child #651
• #650 Hybrid/non-code (stretch)
• #384 Marketplace publication (human Tasks #380, #381, #625, #626)
• #390 Licensing
• #627 Original vision pointer (superseded by this Release for checklist)

See docs/design/pm-orchestrator-architecture-and-browser.md for supporting design (personas prompts, OpenAPI sketch, sequence diagrams, risk mitigations).

Execution plan: Phase 0 hygiene → safety (#657) → feed/planning (#656) → PM (#660) → loops (#638/#639) → adoption (#633) → media (#658) → scheduled ops (#659) → collab → browser (#661) → v1.0 cut. Full roadmap in session plan artifact.

See child issues, #627 (pointer), SPEC.md (v1.0.0 Target State + Endless Feed UX Principles + new End User persona + PM as core + adoption via orchestrator), #631 (detailed feed user stories), #638 (bug loop stories), #639 (feature impl stories), #629 (release planning stories), #641 (release cut/finalize + marketing + combined cycle + refactor + epic implement + release ceremony + PM edges stories/metrics), #642 (scheduled monitoring stories via PM), #660 (PM/Orchestrator with architecture + integrations + labels + webhooks + risks/metrics), #633 (adoption <30m stories with PM orchestration + edges + metrics), #661 (browser UX/API), #652 (marketplace via PM), #643 (human collab depth via PM) for details.

Status: Far-future target under active phased execution (started 2026-07-24). Checklist truth-upped: prior [x] sketch progress demoted until E2E-complete. Hierarchy formally linked. Next: intermediate releases via #612, agent collab bugs, then Phase 1 safety foundations.

Q&A refinements captured (2026-06-17 sessions)

• Structure: #654 is primary Roadmap/Release. Sub-Epics created (#659, #660, #661) and children assigned in text.
• Top priorities for UX refinement: Simulation (#645), Checkpoint primitive (#648), Multi-agent (#644) — concrete example flows added.
• SPEC updates + Endless Feed UX Principles + PM as core + adoption via orchestrator added.
• Detailed stories added across planning, feed, autonomy loops, release ops, scheduled monitoring, refactor, epic implement, adoption (incl. edges), release ceremony, marketplace, human collab, with PM orchestration, personas, checkpoints, budget, metrics.
• User introduced 'Project Manager/Orchestrator' as potential core v1.0.0 deliverable with long-running orchestration, sub-agent personas (devs with personalities, designers with aesthetics, third-party), polling What's Next, unified Q&A API, budget keeping, future browser dashboard.
• Architecture for PM detailed (core loop, assignment, API, coordination, budget/safety, observability, integrations to what_next/#470/MCP/.plate, webhooks, driving labels).
• Browser surface #661 with UX flows/wireframes (incl. driver badges/editing, modals, toasts, GitHub embeds) and API contract.
• Design doc created/expanded with personas examples, OpenAPI, diagrams, risks.
• AGENTS.md updated with orchestration guidance (labels, webhooks, mapping, SOR).
• Adoption #633 now has PM-orchestrated <30m stories + edges + metrics; release ceremony + PM edges in #641; market via PM in #642; human collab depth in #643; PM risks/metrics in #660.

Autonomous execution note (2026-06-18): Questions #327/.plate-root and #341/SPEC-routing now have signals met (close-ready via contemplation; see their comments). Legacy #390 (licensing) and #384 (marketplace) triaged/integrated as v1 adoption prereqs (human merges/pubs pending). See full plan for 5 phases to zero opens + v1.0 cut. Progress: 2 Questions + 2 legacy integrated. #653, #661, #648, #645 progressed with code/tests/fragments/MCP. Cycle run (dry) showed throttling.

Wrap note (this Q&A round): Per user direction, this interactive Q&A refinement round is paused. #654 serves as the current summary of the v1.0.0 vision and all captured refinements. No further immediate changes. Future sessions can resume.

Start with docs/design/pm-orchestrator-architecture-and-browser.md, SPEC.md, and the linked child issues, especially #656–#661. Trace the readiness criteria and execution plan before choosing a narrower child issue; this roadmap is done only when its listed v1.0.0 capabilities have production-complete behavior and E2E evidence.