Back to directory

engrim

A local SQLite engine that stores a coding agent's architectural decisions and project constraints as compact episodic memory, so switching between tools such as Claude Code, Cursor or Codex does not lose prior context.

engrim

engrim is a project-scoped memory engine meant to replace re-sending a full chat transcript on every turn. It stores project decisions, constraints and state in a local SQLite file and retrieves a curated few thousand characters of relevant memory instead of tens of thousands of tokens of past conversation, aiming to reduce both cost and the loss of sharpness that comes from long contexts.

It integrates with several agent hosts, including Google Antigravity, Claude Code, Cursor, Codex CLI, GitHub Copilot CLI, OpenCode and Windsurf, by installing hooks and an MCP server for each. A setup command configures the hooks, a matching skill file, and the MCP server registration for the chosen host, and an integrity check validates the SQLite database and its semantic model.

More in Agent memory

See all 450
  • A long technical guide to AI agent memory: why stateless LLMs need it, the different memory architectures from vector stores to knowledge graphs, and how Cognee implements one as an open-source memory layer.

  • A local, file-based long-term memory layer for a personal AI that never summarizes conversations, keeping every line with a timestamp and searching by time first and words second, rather than retrieving only similar-seeming chunks.

  • NEOTH

    deepwiki.com

    A local-first personal AI daemon in Rust with five memory tiers, consent-gated tools, a WASM plugin sandbox, multi-provider model routing and signed audit logs, aimed at people who want both a daily assistant and a serious coding helper.