- CollectionMemory & State
- Classificationagent-native
- LicenseMIT
- MCPAvailable
- Agent SkillUnavailable
- InterfacesMCP · CLI · STDIO · http · API
- Verified2026-10-13
What it does
MemPalace is a local-first memory system for AI agents. It stores conversation history as verbatim text and retrieves it with semantic search. Official copy: it does not summarize, extract, or paraphrase. The index is a palace metaphor — people and projects are wings, topics are rooms, original content lives in drawers — so searches can be scoped instead of run against a flat corpus.
The retrieval backend is pluggable (ChromaDB default; SQLite exact, Milvus, Qdrant, and pgvector are documented). Nothing leaves the machine unless the operator opts in. A temporal entity-relationship graph with validity windows sits on local SQLite. Each specialist agent can get its own wing and diary, discoverable at runtime via mempalace_list_agents.
Agent-facing interfaces
- Interaction pattern: CLI + MCP
- Catalog protocols: MCP, CLI, STDIO, http, API
- MCP: ✅ Available
- Agent Skill: ⚠️ No official npx skills add package published yet.
- Onboarding type:
mcp
How agents onboard
Interaction pattern: CLI + MCP
uv tool install mempalace
mempalace init ~/projects/myapp
mempalace mine ~/projects/myapp
mempalace search "why did we switch to GraphQL"
mempalace wake-up
See the full onboarding section in the dossier.
Protocol surface
| Interface | Detail |
|---|---|
| CLI | uv tool install mempalace then init, mine, search, wake-up |
| Python API | pip install mempalace inside a venv |
| MCP stdio | Local CLI or docker run -i ghcr.io/mempalace/mempalace |
| Docker | ghcr.io/mempalace/mempalace:latest; data under /data |
| Docs | Getting started, MCP tools |
Verification
Last verified 2026-10-13.
Sources:
- https://api.github.com/repos/MemPalace/mempalace