$ loading_
Provide local-first memory and retrieval for coding agents using SQLite embeddings.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Palace" yet — see the docs or source repo.
When using a coding agent locally, developers can store project background, constraints, and past decisions in this tool for future tasks. It fits workflows that need persistent context while keeping data local.
If your agent stack uses MCP, this tool can act as a memory server and provide a unified read/write interface for coding-task memory. It is useful when memory needs to be modularized in the workflow.
When a team or individual wants to manage agent memory in a single SQLite file, this tool can store and retrieve project-related information. It suits lightweight deployment and local-first data handling.
It is a local-first memory tool for coding agents, exposed as an MCP server and using a single SQLite file with local embeddings. Its core purpose is to store and retrieve long-term context for agents.
The provided information says it uses a single SQLite file for storage and runs as an MCP server. For exact installation and runtime details, see the source repository.
From the description, it emphasizes local-first behavior and local embeddings, suggesting local execution and local data retention. The provided material does not specify broader remote capabilities; see the source repository.
Build local codebase memory for AI agents with search and architecture inspection.
Give AI coding agents persistent local memory across sessions.
Share, search, and reuse local memory across multiple AI coding agents.
Give AI coding assistants persistent, conflict-aware memory across projects and sessions.
Build semantic memory and structural code indexes for persistent AI project context.
Provide persistent local semantic memory for MCP tools to store and search notes.