bytelyst/learning_ai_common_plat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs: add phase 1 ecosystem execution plan

Browse Source
This commit is contained in:
Saravana Achu Mac 2026-04-03 16:21:04 -07:00
parent c97e693a7d
commit 3d59e7bbed
3 changed files with 246 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
docs/ecosystem/ECOSYSTEM_IMPLEMENTATION_TRACKER.md
View File

@ -137,3 +137,11 @@ Example:
- [ ] contract-test strategy needs fixture location and validation tooling decisions
These should be resolved before claiming the ecosystem docs are fully implementation-ready.
---
## 5. Active Execution Plan
Current implementation kickoff document:
- [`PHASE1_TRANSCRIPT_NOTE_MEMORY_EXECUTION_PLAN.md`](/Users/saravana/BytelystAI/learning_ai/learning_ai_common_plat/docs/ecosystem/PHASE1_TRANSCRIPT_NOTE_MEMORY_EXECUTION_PLAN.md)

237
docs/ecosystem/PHASE1_TRANSCRIPT_NOTE_MEMORY_EXECUTION_PLAN.md Normal file
View File

@ -0,0 +1,237 @@
# Phase 1 Execution Plan: Transcript -> Note -> Memory
> **Status:** Ready for implementation
> **Owner:** `learning_ai_common_plat`
> **Golden Flow:** LysnrAI transcript -> NoteLett note -> MindLyst memory
> **Primary Leads:** Codex for shared-contract and sequencing work, Saravana for final approval and prioritization
---
## 1. Scope
This phase implements one end-to-end ecosystem flow:
1. a transcript is created in `learning_voice_ai_agent`
2. that transcript is represented as a shared artifact
3. `learning_ai_notes` consumes it and creates or links a note artifact
4. `learning_multimodal_memory_agents` consumes the transcript/note lineage and creates or proposes a memory artifact
5. canonical events and provenance are preserved across the chain
This phase is intentionally narrow. It is the first implementation proof for the shared ecosystem contracts.
---
## 2. Repos In Scope
- [`learning_ai_common_plat`](/Users/saravana/BytelystAI/learning_ai/learning_ai_common_plat)
- [`learning_voice_ai_agent`](/Users/saravana/BytelystAI/learning_ai/learning_voice_ai_agent)
- [`learning_ai_notes`](/Users/saravana/BytelystAI/learning_ai/learning_ai_notes)
- [`learning_multimodal_memory_agents`](/Users/saravana/BytelystAI/learning_ai/learning_multimodal_memory_agents)
Out of scope for this phase:
- ActionTrail replay UI
- auth approvals for the flow
- timeline UI
- marketplace work
---
## 3. Implementation Order
Strict order:
1. harden shared contracts in `learning_ai_common_plat`
2. implement producer side in `learning_voice_ai_agent`
3. implement note consumer/producer in `learning_ai_notes`
4. implement memory consumer/producer in `learning_multimodal_memory_agents`
5. run cross-repo verification and update tracker
Do not start product implementation before contract fields and event names are stable enough for Phase 1.
---
## 4. Shared Contract Tasks
Repo: [`learning_ai_common_plat`](/Users/saravana/BytelystAI/learning_ai/learning_ai_common_plat)
### 4.1 Artifact contract hardening
- [ ] finalize the transcript artifact envelope
- [ ] finalize the note artifact envelope expectations
- [ ] finalize the memory artifact envelope expectations
- [ ] define required provenance fields for the flow
- [ ] define `artifact.linked` relation usage for transcript -> note and note -> memory
### 4.2 Event taxonomy hardening
- [ ] finalize `capture.transcript.created`
- [ ] finalize `artifact.created`
- [ ] finalize `artifact.linked`
- [ ] finalize `memory.entry.created`
- [ ] add one canonical example payload per event
### 4.3 Verification/fixture prep
- [ ] decide fixture location for Phase 1 examples
- [ ] create canonical JSON examples for transcript/note/memory artifacts
- [ ] create canonical JSON examples for the 4 phase-1 events
Commits:
- pending
---
## 5. Repo Task Breakdown
### 5.1 learning_voice_ai_agent
Repo: [`learning_voice_ai_agent`](/Users/saravana/BytelystAI/learning_ai/learning_voice_ai_agent)
Implementation tasks:
- [ ] identify the transcript creation boundary to hook artifact emission
- [ ] map transcript output into the shared artifact envelope
- [ ] emit `capture.transcript.created`
- [ ] attach correlation/session identifiers for downstream consumers
- [ ] expose or persist the artifact in a way downstream consumers can access
Verification:
- [ ] transcript artifact matches canonical fixture
- [ ] event payload matches canonical fixture
- [ ] provenance fields are populated
Commits:
- pending
### 5.2 learning_ai_notes
Repo: [`learning_ai_notes`](/Users/saravana/BytelystAI/learning_ai/learning_ai_notes)
Implementation tasks:
- [ ] consume transcript artifact reference or payload
- [ ] create a linked note artifact using the shared envelope
- [ ] emit `artifact.created`
- [ ] emit `artifact.linked`
- [ ] preserve provenance to the originating transcript artifact
Verification:
- [ ] note artifact matches canonical fixture
- [ ] link relation is present and correct
- [ ] no transcript provenance is lost
Commits:
- pending
### 5.3 learning_multimodal_memory_agents
Repo: [`learning_multimodal_memory_agents`](/Users/saravana/BytelystAI/learning_ai/learning_multimodal_memory_agents)
Implementation tasks:
- [ ] consume transcript/note lineage
- [ ] create or propose a memory artifact using the shared envelope
- [ ] emit `memory.entry.created`
- [ ] preserve lineage to transcript and note artifacts
Verification:
- [ ] memory artifact matches canonical fixture
- [ ] lineage points back to upstream artifacts
- [ ] event payload matches canonical fixture
Commits:
- pending
---
## 6. Recommended Commit Sequence
Use this sequence unless a better dependency split is discovered:
1. `learning_ai_common_plat`
- contract hardening
2. `learning_ai_common_plat`
- fixture/examples and tracker updates
3. `learning_voice_ai_agent`
- transcript artifact + event emission
4. `learning_ai_notes`
- note artifact + linking
5. `learning_multimodal_memory_agents`
- memory artifact + lineage
6. `learning_ai_common_plat`
- final verification/tracker commit
This keeps downstream repos aligned to a stable contract before product work lands.
---
## 7. Verification Commands
The exact commands may evolve by repo, but the evidence should include at least:
### learning_ai_common_plat
- [ ] schema/examples validated locally
- [ ] docs/tracker updated
### learning_voice_ai_agent
- [ ] relevant backend or product tests covering transcript creation path
- [ ] fixture conformance evidence recorded
### learning_ai_notes
- [ ] relevant backend/web tests covering artifact creation/linking path
- [ ] fixture conformance evidence recorded
### learning_multimodal_memory_agents
- [ ] relevant backend tests covering memory creation/proposal path
- [ ] fixture conformance evidence recorded
### Cross-repo
- [ ] one end-to-end golden flow walkthrough recorded in the tracker
- [ ] commit hashes added for all participating repos
- [ ] residual risks captured
---
## 8. Done Criteria
This phase is done only when all of the following are true:
- [ ] transcript artifact exists in canonical shape
- [ ] note artifact exists in canonical shape
- [ ] memory artifact exists in canonical shape
- [ ] required phase-1 events are emitted in canonical form
- [ ] provenance and lineage are preserved
- [ ] tracker has commit links for all participating repos
- [ ] residual gaps are documented
---
## 9. Known Risks
1. Product repos may have different persistence models that make the artifact handoff uneven.
2. Event emission may be easier than downstream retrieval, so transport or lookup assumptions must stay explicit.
3. Memory creation may need a “proposed” intermediate state rather than immediate durable creation.
4. If the phase-1 event set changes mid-implementation, product repos can drift quickly.
---
## 10. Next Step After Phase 1
If this flow succeeds, the next recommended implementation flow is:
- FlowMonk plan -> ChronoMind routine -> EffoRise habit
That would validate a second shared seam: planning/time/habits rather than capture/knowledge/memory.

1
docs/ecosystem/README.md
View File

@ -24,6 +24,7 @@ Recommended structure:
- [`ECOSYSTEM_CONTRACT_TEST_STRATEGY.md`](/Users/saravana/BytelystAI/learning_ai/learning_ai_common_plat/docs/ecosystem/ECOSYSTEM_CONTRACT_TEST_STRATEGY.md) — how to prevent contract drift across repos
- [`ECOSYSTEM_MARKETPLACE_UNIFICATION.md`](/Users/saravana/BytelystAI/learning_ai/learning_ai_common_plat/docs/ecosystem/ECOSYSTEM_MARKETPLACE_UNIFICATION.md) — shared marketplace model
- [`ECOSYSTEM_PERSONAL_TIMELINE_PRD.md`](/Users/saravana/BytelystAI/learning_ai/learning_ai_common_plat/docs/ecosystem/ECOSYSTEM_PERSONAL_TIMELINE_PRD.md) — unified user activity stream
- [`PHASE1_TRANSCRIPT_NOTE_MEMORY_EXECUTION_PLAN.md`](/Users/saravana/BytelystAI/learning_ai/learning_ai_common_plat/docs/ecosystem/PHASE1_TRANSCRIPT_NOTE_MEMORY_EXECUTION_PLAN.md) — execution-ready plan for the first golden flow
- [`adoption/`](/Users/saravana/BytelystAI/learning_ai/learning_ai_common_plat/docs/ecosystem/adoption) — per-product adoption notes
Ownership model: