Engraphis is a local-first memory engine for AI coding agents. It addresses the problem of agents forgetting by maintaining durable, scoped, and explainable memories stored in a single SQLite file on the user's machine.
The tool includes retention mechanics that cause memory to fade without recall while reinforcement makes memories persist. Its dashboard WebUI, launched with one command, provides 11 tabs with no cloud or signup required. These include an overview showing memory counts, retention distribution, weekly growth, and decay forecast; hybrid search with score breakdowns; a memories browser organized by workspace and sorted by retention; an interactive force-directed knowledge graph of entities and relationships; on-demand consolidation sweeps with distillation results; a chat interface for asking questions grounded in stored context; import via drag-and-drop or pasted markdown files; and views for timeline, audit, settings, and vaults along with bi-temporal history.
It runs entirely on the local machine and works with Claude Code, Cursor, Cline, Zed, and Windsurf. Installation uses a pip command that also launches the dashboard at http://127.0.0.1:8700. The project is free and open-source under the Apache-2.0 license at its core, with version 0.9.0 available.
Engraphis sits in PulseGate's AI & ML category. AI coding agents forgetting important context, relationships, and history across sessions. It is built as an open-source project for developers. The project is open source (Apache-2.0). It runs on the web and the command line, and it can be self-hosted.
Engraphis first shipped in 2026. The project is developed in the open on GitHub with 126 stars and 256 commits in the last 90 days. Key capabilities include Knowledge Graph, Hybrid Search, and Memory Consolidation. It exposes integrations via a public API.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do