Merge branch 'main' of https://kevinblog.sytes.net/Code/Jaked/JiboExperiments

OpenJibo/docs/feature-backlog.md

@@ -614,6 +614,8 @@ Current release theme:
   - recognition, enrollment, rename, and profile-correction boundaries
   - split between local state and hosted cloud state
   - first useful hosted identity slice
+  - live QA has shown person-identification collisions in the same loop (for example, a parent and child both getting normalized to the same remembered name)
+  - person-identification correction likely needs its own repair pass before we can trust greetings, reports, and presence triggers in mixed-household scenarios
 
 ### 20. Onboarding, Loop Management, And Fresh Start
 
@@ -638,9 +640,12 @@ Current release theme:
   - `make a pizza` now ports the original scripted-response path through `chitchat-skill` with `mim_id = RA_JBO_MakePizza` and pizza-making animation ESML
   - `can you order pizza` now ports the original scripted-response path through `chitchat-skill` with `mim_id = RA_JBO_OrderPizza`
   - current source answers these with a `1.0.19` rule-based persona baseline, backed by `OpenJiboCloudBuildInfo.PersonaBirthday`
+  - `how old are you` now also uses the imported Build B age prompts so the first-powered-up and birthday phrasing stays source-backed
 - Follow-up:
   - wire persona age to first-powered-up or durable first-cloud-seen metadata when available
   - add command-vs-question variants so expressive prompts can answer conversationally before launching actions
+  - live QA has shown motion/sleep quirks too: `turn around` can become a no-op and `go to sleep` can fail at the last step before the sleep animation fully completes
+  - reply-selection polish still needs attention on a couple of identity prompts where short variants are over-selected (`how are you`, `what is your favorite flower`)
 
 ### 22. Command Vs Question Reply Style
 
@@ -770,7 +775,7 @@ Current release theme:
 
 ### 28. Grocery List Capability (Requested Feature)
 
-- Status: `discovery`
+- Status: `in_progress`
 - Tags: `content`, `docs`, `storage`
 - Why now:
   - directly requested by Jibo owners and fits memory + household utility roadmap
@@ -779,13 +784,14 @@ Current release theme:
   - examples:
     - `C:\Projects\jibo\pegasus\packages\chitchat-skill\mims\scripted-responses\RA_JBO_ShoppingList.mim`
     - `C:\Projects\jibo\pegasus\packages\chitchat-skill\mims\scripted-responses\RA_JBO_ManageToDoList.mim`
-- Candidate delivery paths:
-  - native lightweight list skill (fastest user value)
-  - integration-backed list orchestration (long-term richer ecosystem fit)
+- MVP decision:
+  - use the existing household list engine as the native lightweight grocery MVP
+  - keep grocery as a first-class spoken alias over the shopping list storage path
+  - reserve integration-backed list orchestration for a later discovery pass
 - Exit criteria:
-  - clear decision on MVP path
-  - first schema for list items + ownership scope
-  - initial voice flows and follow-up intent handling defined
+  - grocery prompts, add/recall/done flows, and list follow-ups consistently speak grocery wording
+  - existing shopping/to-do flows remain unchanged
+  - future integration-backed list work remains a separate backlog item
 
 ### 29. Legacy MIM Personality Import Ladder
 
@@ -889,7 +895,10 @@ Current release theme:
   - richer identity follow-ups like `who is this`, `do you know me`, `do you remember me`, and `can you recognize me`
   - mood and affect prompts like `how are you`, `are you happy`, `are you sad`, and `are you angry`
   - self-description charm like `what's your name`, `do you have a nickname`, `do you like being Jibo`, and `what is your favorite name`
+  - deeper personality follow-ups like `what do you dream about`, `what are you afraid of`, `what do you want to talk about`, `what is your best book`, `what is your best exercise`, `what is your dream vacation`, `who is your hero`, `who do you love`, and `what is your religion`; `what is your sign` stays deferred until templated placeholder rendering exists
+  - the next identity / knowledge wave adds `are you god`, `are you here`, `do you have super powers`, `how much do you know`, `what does jibo mean`, `where do you get info`, `what are you forbidden to do`, `what color are you`, and `what do you do when alone`
   - additional legacy source-backed `RI_USR` prompts where the text is short and the behavior is easy to verify
+  - templated edge cases like `what is your sign`, `how many people do you know`, and `what is the loop` where live birthday and loop state are part of the line instead of a plain canned response
 - Exit criteria:
   - a stable checklist exists for the original persona surface
   - each pass can be scoped to a small batch of prompts
@@ -975,7 +984,9 @@ For `1.0.19`:
   - next implementation pass should supply the real Azure Storage connection string / deployment wiring and validate the live round-trip in the storage account smoke test
 10. Update, backup, and restore proof - implemented (update creation and backup creation now survive persisted reloads; restore is the persisted-state rehydration proof path, not a new cloud API)
 11. STT upgrade and noise screening
+  - progress update (`2026-05-21`): added a low-signal short-turn screen in websocket finalization so filler-only fragments and stray single-token leftovers like `so command` get rejected before they can become bad turns, while preserving the existing yes/no and word-of-the-day short-turn flows
 12. Hosted capture/storage plan / indexing for group testing
+  - progress update (`2026-05-21`): added a bundle helper so group testers can package raw capture trees, `capture-index.ndjson`, and exported fixtures into one zip handoff artifact
 13. Binary-safe media storage / sync to cloud drive: OneDrive, Google Drive, Box, etc.
 14. Provider-backed news and weather parity polish
 15. Grocery list capability discovery and MVP selection

OpenJibo/docs/live-jibo-capture.md

@@ -77,3 +77,18 @@ Useful helper scripts:
 - [scripts/cloud/get-websocket-capture-summary.sh](/OpenJibo/scripts/cloud/get-websocket-capture-summary.sh)
 - [scripts/cloud/import-websocket-capture-fixture.py](/OpenJibo/scripts/cloud/import-websocket-capture-fixture.py)
 - [live-jibo-test-runbook.md](/OpenJibo/docs/live-jibo-test-runbook.md)
+
+## Group Testing Handoff
+
+When you have a useful capture set and want to share it with another tester, bundle the capture root into a single zip so the raw events, capture index, and exported fixtures stay together.
+
+Recommended helper:
+
+- [scripts/cloud/New-CaptureBundle.ps1](/OpenJibo/scripts/cloud/New-CaptureBundle.ps1)
+
+The bundle includes:
+
+- `capture-index.ndjson`
+- websocket and HTTP `*.events.ndjson` files
+- exported `*.flow.json` fixtures
+- a small `bundle-manifest.json` with file counts and source metadata

OpenJibo/docs/release-1.0.19-plan.md

@@ -6,6 +6,8 @@ This release starts the shift from `1.0.18` hardening to visible feature growth.
 
 The goal is to keep compatibility work steady while shipping personality and capability slices that make OpenJibo feel less like a placeholder cloud and more like a real assistant platform.
 
+For grocery list capability, the 1.0.19 MVP choice is the existing household list engine with grocery as a first-class spoken alias. That keeps the storage model simple now while leaving integration-backed list orchestration for a later pass.
+
 ## Snapshot
 
 - Kickoff date: `2026-05-05`
@@ -56,6 +58,13 @@ Current batch note:
 - the favorites batch now includes `what is your favorite animal`, `what is your favorite bird`, `do you like penguins`, and `do you like animals` so the penguin-centered replies stay close to Pegasus
 - the latest social batch adds `welcome back`, `what are you thinking`, `what have you been doing`, and `what did you do` so presence and charm stay lively without distracting from the memory roadmap
 - the newest identity-charm batch adds `what's your name`, `do you have a nickname`, `do you like being Jibo`, `are there others like you`, and `what is your favorite name` so the robot stays familiar while still sounding like Pegasus
+- the next deep-personality batch adds `what do you dream about`, `what are you afraid of`, `what do you want to talk about`, `what is your best book`, `what is your best exercise`, `what is your dream vacation`, `who is your hero`, `who do you love`, and `what is your religion`; `what is your sign` is still deferred until we add templated placeholder rendering
+- the next identity/knowledge batch adds `are you god`, `are you here`, `do you have super powers`, `how much do you know`, `what does jibo mean`, `where do you get info`, `what are you forbidden to do`, `what color are you`, and `what do you do when alone`
+- the next body/mission batch adds `how much do you weigh`, `how tall are you`, `how much do you cost`, `what if I unplug you`, `what is your purpose`, `what is your prime directive`, `what is jibo commander`, `do you like commander app`, and `what are you made of`
+- the templated edge-case batch adds `what is your sign`, `how many people do you know`, and `what is the loop` so the remaining source-backed lines can lean on live birthday and loop state
+- the work/eat/home batch adds `how do you work`, `what do you eat`, `where do you live`, and `what languages do you speak` so the everyday self-description cluster keeps moving toward the original phrasing
+- the age batch adds `how old are you` through `JBO_HowOldAreYou` so the birthday and first-powered-up phrasing stays source-backed instead of falling back to a generic age answer
+- live QA has surfaced a few repair targets to carry into the next pass: person-identification collisions inside the same loop, `turn around` / `go to sleep` motion quirks, and a couple of reply-selection spots where short variants are being over-selected (`how are you`, `what is your favorite flower`)
 - this pass keeps Build B moving while still favoring source-backed phrasing and preserving the command-vs-question boundary
 - the next passes should keep the same pattern and prefer source-backed phrasing whenever the legacy MIM text is available
   - if a source-backed legacy line is missing, use a temporary direct reply only to keep the pass moving, then backfill source text later
@@ -84,6 +93,8 @@ The goal is to port these in small batches, capture the source-backed phrasing w
 - the restore proof is the persisted-state rehydration path; do not scope it into a new hosted restore API until we have real device evidence
 - continue alarm/gallery/yes-no cleanup from `1.0.18` evidence where regressions are still open
 - improve short-turn STT reliability and low-signal screening
+- the latest STT pass adds a websocket-side low-signal screen for filler-only and stray single-token leftovers while keeping yes/no and word-of-the-day turns intact
+- capture indexing and group-test handoff now have a bundle helper that packages raw event captures, the index manifest, and exported fixtures together for easier review/share flows
 
 ### 3. Pegasus-To-Cloud Platform Porting