* [GrayMatter v0.15.0](https://github.com/angelnicolasc/graymatter) – Three-line embeddable agent memory layer providing persistent memory and reducing token usage by about 90%.

* [deja-vu v0.19.0](https://github.com/vshulcz/deja-vu) – Zero-dependency CLI memory layer that indexes Claude Code, Codex, and opencode session logs for fast search, agent recall, redaction, stats, and sync.

* [Flatbread v1.1.0](https://github.com/FlatbreadLabs/flatbread) – Turns Git-backed Markdown and YAML files into a typed relational graph queried via GraphQL, generated TypeScript, or the Flatbread CLI.

* [brain.md v0.2.0](https://github.com/mindmuxai/brain.md) – A repo-native, agent-agnostic persistent memory layer that captures durable project decisions as Markdown via a zero-dependency CLI.

* [Mnemon v0.2.0](https://github.com/mnemon-dev/mnemon) – LLM-supervised persistent memory for AI agents with graph-based recall, intent-aware retrieval, importance decay, and automatic deduplication.

* [Wife v1.0.0](https://github.com/ma-nucho-pro/wife) – Local, persistent and auditable memory layer that stages facts until confirmed across sessions to keep agents consistent.

* [Signet AI v0.157.0](https://github.com/Signet-AI/signetai) – Local-first identity, memory, and secret management for AI agents, providing portable state across sessions, models, and harnesses.

* [ax v0.43.0](https://github.com/Necmttn/ax) – Local agent-experience loop that learns from AI coding agent sessions, records retro experiments in a typed graph, and proposes repo-specific fixes for review.

* [Ogham MCP v0.19.0](https://github.com/ogham-mcp/ogham-mcp) – Persistent, searchable shared memory server for AI coding agents that syncs across clients.

* [hypha v1.0.1](https://github.com/CodeSoul-co/Hypha) – TypeScript framework combining an agent core with a production harness to run governed, durable domain agents with checkpoints, recovery, and cache reuse.

* [AtomicMemory cli-v0.2.0](https://github.com/atomicstrata/atomicmemory) – Inspectable, portable semantic memory layer for agents and applications, supporting portable capture, groundings, and correction-aware knowledge across SDK, CLI, MCP, adapters, and plugins.

* [Bitterbot orchestrator-v0.2.2](https://github.com/Bitterbot-AI/bitterbot-desktop) – Local-first personal AI with biologically inspired memory, a dream engine, and a P2P skills marketplace.