Skip to content

v0.4.0 – Bridge scope clarified, memory engineering moved out

Latest

Choose a tag to compare

@apajon apajon released this 24 Apr 15:46
· 2 commits to main since this release
a49f4b8

This release clarifies the scope of mempalace-mcp-bridge as the operational bridge for MemPalace with VS Code Copilot Chat via MCP.

What changed:

  • advanced structured memory methodology has been moved out to the dedicated mempalace-memory-engineering repository
  • bridge-side methodology docs now redirect readers to the new canonical repository
  • the README has been reworked to keep the focus on setup, runtime integration, VS Code Copilot MCP configuration, verification, troubleshooting, and devcontainer usage
  • documentation was reconciled with main so the bridge-facing onboarding and limitations remain explicit

Why this matters:

This repository had started to mix two different concerns:

  • the bridge itself: setup, MCP wiring, runtime verification, compatibility handling, and troubleshooting
  • memory engineering methodology: wings and rooms, retrieval order, deduplication, persistence rules, and instruction patterns

Those concerns now evolve separately. This release keeps the bridge repository focused on operating MemPalace reliably inside VS Code Copilot Chat, while moving the broader methodology to its own dedicated home.

Scope of this release:

  • clearer repository boundary around the bridge/runtime layer
  • retained focus on VS Code Copilot Chat MCP integration
  • redirect stubs for moved methodology documents
  • updated README and supporting documentation structure

What did not change:

  • supported stable bridge target remains ChromaDB 0.6.x
  • setup, update, verify, and MCP launcher behavior are unchanged
  • devcontainer usage remains supported
  • reconstruction and investigation tooling remain in the repository
  • no new runtime feature or compatibility change is introduced by this release

Known limitations:

  • the stable bridge path still targets ChromaDB 0.6.x, not 1.x
  • non-0.6.x palaces remain blocked in the stable path
  • experimental 1.x work is still outside the supported main bridge workflow
  • Linux / WSL2 remains the tested path

Canonical methodology repo: