Bump version to 2.35.0: Knowledge Resilience & Standardization Milestone.

2026-05-02 13:18:07 +09:00
parent 11f25534d0
commit 8bb8c065d7
27 changed files with 2452 additions and 14 deletions
@@ -0,0 +1,58 @@
+# Feature Plan: Project Chronicle Guard
+
+## 1. Feature Name
+Project Chronicle Guard
+
+## 2. Reason
+The user wants AI-human planning, questions, decisions, implementation details, bugs, and retrospectives to remain as project-specific Markdown knowledge instead of disappearing into chat history.
+
+## 3. Original User Request
+Add a Designer menu to the sidebar, similar in spirit to Team, and implement a staged Project Chronicle Guard system that records planning, decisions, development logs, bug fixes, and retrospectives.
+
+## 4. Interpreted Intent
+The user does not only want another chat prompt preset. They want a durable project knowledge layer that captures why decisions were made and keeps records separated by project.
+
+## 5. Problems To Solve
+- Later readers forget why a direction was chosen.
+- AI questions lose their intent and decision impact.
+- Planning and implementation history are mixed or absent.
+- Multiple projects can contaminate each other's records.
+
+## 6. MVP Scope
+- Add an independent `projectChronicle` module.
+- Add project profile types and Markdown writer utilities.
+- Add sidebar Designer controls for record project selection and basic project management.
+- Generate the project folder structure and seed Markdown files.
+- Save a planning document, discussion summary, decision record, development log, bug log, and timeline entry through explicit user actions.
+
+## 7. Out Of Scope
+- Full automatic transcript recording
+- External database storage
+- Git auto-commit
+- Real-time file watching
+- Search UI
+- Tight integration with agent execution internals
+
+## 8. Development Direction
+Start with a low-risk independent module and connect the sidebar through message events. The chat agent can remain unaware of the record module in the first stage.
+
+## 9. Existing Feature Relationship
+The current sidebar already has Brain and Agent controls. The Designer menu should follow that UI pattern but manage project record profiles instead of agent skill files.
+
+## 10. Dependency Strategy
+Use TypeScript, Node `fs`, and Markdown files only. Store project profile metadata in VS Code global state and avoid modifying core chat execution.
+
+## 11. Expected Value
+- Project-specific planning and decision history becomes reusable.
+- The user can intentionally create records before or after development work.
+- Future automation can build on a stable local file structure.
+
+## 12. Success Criteria
+- A Designer selector appears in the sidebar.
+- A new record project can be created.
+- Record folders and seed files are generated safely.
+- Markdown documents can be generated without needing external services.
+- Existing chat, brain, and agent flows continue to compile.
+
+## 13. Developer Instruction
+Implement the first stage only: independent module, basic Designer menu, project creation/selection, folder seeding, and explicit sample record generation actions. Keep automatic conversation analysis for a later stage.
@@ -0,0 +1,36 @@
+# Feature Plan: Second Brain Trace Mode
+
+## 1. Feature Name
+Second Brain Trace Mode
+
+## 2. Reason
+The user wants to verify whether AI answers actually use Second Brain notes, which documents were searched, and how retrieved knowledge influenced the answer.
+
+## 3. Original User Request
+Add a way to see if the AI searched Second Brain, which notes it referenced, whether each note was used in the final answer, and why unused notes were excluded.
+
+## 4. Interpreted User Intent
+The user does not want to rely on vague claims that the AI used memory. They want visible grounding evidence and a developer debug trace.
+
+## 5. Scope
+- Add a trace mode toggle in the sidebar.
+- Add a debug JSON toggle.
+- Retrieve relevant Markdown notes from the active Second Brain.
+- Inject selected notes into the answer context.
+- Append a user-visible 2nd Brain usage section to answers.
+- Show searched-but-unused notes and basic grounding metrics.
+
+## 6. Out Of Scope
+- Vector database integration
+- Full semantic embedding search
+- Perfect factual attribution at sentence level
+- External observability tools
+
+## 7. Development Direction
+Use a low-dependency local Markdown search first. Keep vector search and richer grounding analysis as later expansion.
+
+## 8. Success Criteria
+- Users can see whether Second Brain was used.
+- Users can see referenced Markdown paths.
+- Debug mode exposes retrieval JSON.
+- General questions can explicitly show that Second Brain was not used.