What problem it solves

AI coding sessions and project context can be difficult to retain and recover across tools and sessions. MemPalace addresses this by storing source content locally without summarizing or paraphrasing it, then making that content searchable through a structured, scope-aware retrieval system.

Key features

• Verbatim semantic memory — Conversation history and project content are stored as original text and retrieved through semantic search without summaries, extraction, or paraphrase.

  Source: README

• Structured, scoped retrieval — People and projects are organized as wings, topics as rooms, and original content as drawers, allowing searches to be scoped rather than performed against a flat corpus.

  Source: README

• Pluggable storage backends — The default retrieval backend is ChromaDB, and alternative backends can be added through a defined interface without changing the rest of the system.

  Source: README, mempalace

• MCP and agent coordination — The MCP interface covers palace reads and writes, knowledge-graph operations, cross-wing navigation, drawer management, agent diaries, and coordination through logstream events and artifact handoffs.

  Source: README

• Local setup and session retention — The project supports guided agent setup, isolated CLI installation, Docker deployment, and auto-save hooks for supported coding tools. Its knowledge graph also supports time-bounded entity relationships.

  Source: README, hooks

Quick start

Copied verbatim from the project's README.

uv tool install mempalace
mempalace init ~/projects/myapp

# Mine content into the palace
mempalace mine ~/projects/myapp                    # project files
mempalace mine ~/.claude/projects/ --mode convos   # Claude Code sessions (scope with --wing per project)

# Search
mempalace search "why did we switch to GraphQL"

# Load context for a new session
mempalace wake-up

# Score how well organized the palace is (read-only, safe while the MCP server runs)
mempalace audit

Who it may suit

This may suit developers and AI coding-tool users who need local, searchable memory across projects and conversations.

Topics to discuss

• How should verbatim local memory balance privacy, relevance, and retrieval speed?

• Which additional storage backends would be most useful to maintainers and users?

• How should MCP tools and agent coordination workflows evolve as memory features expand?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/MemPalace/mempalace#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.