plan: dirgha end-to-end product launch (master)#63
Open
salikshah wants to merge 1 commit into
Open
Conversation
Master plan that ties together every workstream the user described:
login → chat → IDE → CLI → agents → Bucky → marketplace → web platform.
Sections:
1. **The user journey** — A through J: signup, /app, prompt, VPS
provision, dirgha-in-VPS agent, multi-model SQLite, SSE stream,
live IDE edits, session end, Bucky settles, provider payout.
2. **Current state vs target state** — table per journey step.
Big missing piece is step E: gateway today doesn't run dirgha in
VPS for chat sessions. That's the bridge we build in Phase 3.
3. **Six phases, ordered by dependency:**
- Phase 0: in-flight CLI work (ESC fix, scroll-to-top) — days
- Phase 1: CLI agent-DB foundation (7 sprints from agent-db.md)
- Phase 2: Bucky marketplace MVP (items 1-4)
- Phase 3: chat → VPS → dirgha bridge (the integration glue)
- Phase 4: marketplace UX (web)
- Phase 5: login → chat → IDE polish
- Phase 6: demo + VC pitch
~7 weeks serial, ~4-5 weeks with parallelism.
4. **Sequencing rationale** with explicit parallelism markers.
Phases 1 and 2 run concurrently. Phase 3 picks up after both.
Phase 4 picks up after Phase 2.
5. **Dispatch fleet plan** — four waves, A through K agents:
- Wave 1: ESC + sqlite-vec + graph + Bucky MVP (4 in parallel)
- Wave 2: transactional API + bridge (2 agents)
- Wave 3: embeddings + memory/KB + marketplace UX (3 agents)
- Wave 4: polish + demo (2 agents)
6. **Ship gates per phase** — concrete user-facing checks.
E.g. Phase 3 done = browser chat shows `[via VPS:abc]` header
proving the streamed response came through the bridge.
7. **What this plan is NOT** — explicit non-goals to keep scope.
8. **Risks + mitigations** — isolated worktrees, TTFT cap from
v1.20.36, ship-gate checks block merge.
9. **Sub-plans** — bite-sized-task plans for Phases 2-5 to be
drafted after master plan is approved. Phase 1 sub-plan already
exists at docs/cli/index/agent-db.md.
10. **Decision points** — three questions for the user before any
agent fires:
a. Is the order right?
b. Aggressive parallelism (Wave 1 = 4 agents) or staged?
c. Auto-merge after CI green, or manual review?
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
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
1 participant
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.
Master plan tying together login → chat → IDE → CLI → agents → Bucky → marketplace. Six phases, dispatch fleet of A-K agents, ship gates per phase. See
docs/superpowers/plans/2026-05-08-dirgha-end-to-end.md.🤖 Generated with Claude Code