OmniMem

Security Scans / Bandit SAST (push) Successful in 7s

Security Scans / Dependency Audit (mcp_server) (push) Successful in 40s

Security Scans / Dependency Audit (rss_worker) (push) Successful in 30s

Security Scans / Dependency Audit (web_ui) (push) Successful in 27s

Security Scans / Test Coverage (push) Successful in 1m23s

Security Scans / Secret Scan (push) Successful in 4s

Docker Build & Push / Build mcp_server (push) Successful in 17m57s

Docker Build & Push / Build web_ui (push) Successful in 15m23s

Docker Build & Push / Build rss_worker (push) Successful in 2m5s

Every Claude Code session starts from zero. No memory of your project. No memory of what failed last week. No memory that you spent three hours last Tuesday discovering whyonnxruntimeexplodes on Alpine before finding something that actually works.

So you explain the project again. Claude suggests the same broken library again. Same alarm. Same song. You are Bill Murray and Claude is Punxsutawney.

OmniMem fixes that. It is a self-hosted MCP server that gives your AI agent persistent memory across sessions, projects, and machines. It runs on your own hardware and it is free forever.

claude> use onnxruntime for the embeddings ⚠ WARNING: previously abandoned approach onnxruntime — SIGILL crash on Alpine musl libc (effort: 4/5) → switched to sentence-transformers instead

That warning came from memory, not luck. The mistake you already paid for does not get to charge you twice.

One command gets the full stack running from the pre-built Docker Hub images:

curl -fsSL https://codeberg.org/ric_harvey/omnimem/raw/branch/main/install.sh | bash

Five kinds of memory, all searched together at recall time:

The top recall result might be a decision from six months ago on a different project, a solution from yesterday, or an article that landed on Tuesday night. It does not matter where it came from as long as it is useful.

Not just a key-value store with an MCP wrapper. OmniMem models how memory actually works: things fade over time, they sometimes contradict each other, and the hard-won stuff earns its place.

The ranking formula behind every recall:

score = similarity x surface_score x recency x experience_weight

Four factors decide what comes back. Semantic similarity alone is not enough — lifecycle state, age, and how hard the lesson was to learn all play a role.

Four containers. Nothing leaves your machine. Local embeddings via sentence-transformers, storage in Valkey with vector search, and both front doors share the same memory engine.

No SaaS. No vendor lock-in. No context shipped to someone else's servers.

- Valkeyis an open source Redis fork. All your data stays in a named Docker volume on your own machine.
- Multi-arch Docker imagesfor amd64 and arm64. It runs on a Raspberry Pi, AWS Graviton, or Apple Silicon just as well as x86.
- sentence-transformersruns embeddings locally with no API calls.
- MIT licensedmeans fork it, extend it, run it wherever you want.
- One backup commandcallsdump_to_file()and exports everything to a JSON file you own.

The installer checks Docker is installed, generates secure passwords, writes a sensible.env, and starts everything from the pre-built Docker Hub images:

curl -fsSL https://codeberg.org/ric_harvey/omnimem/raw/branch/main/install.sh | bash

git clone https://codeberg.org/ric_harvey/omnimem.git && cd omnimem cp .env.example .env # Set VALKEY_PASSWORD and ANTHROPIC_API_KEY in .env docker compose up -d

Then point your agent at it —Claude Code(~/.claude.json):

{ "mcpServers": { "omnimem": { "type": "sse", "url": "http://localhost:8765/sse" } } }

The server delivers its usage guide to any connecting agent automatically via the MCPinstructionsfield — no configuration file needed. The web dashboard is athttp://localhost:8080.

MIT. Free to use, fork, and modify. No enterprise tier, no hosted version, no strings.

AI trading memory layer for MT5/forex with 15 MCP tools — store/recall trades, pattern discovery, strategy evolution, and Outcome-Weighted Memory.

Hosted MCP server for persistent, shared AI agent memory

Two-layer memory for AI agents with an immune system — episodes compress into identity through citation-validated graduation. Zero dependencies, 5 MCP tools.

A lightweight MCP server for semantic memory storage, knowledge graphs, and cross-session context

A production-ready Model Context Protocol (MCP) server that provides a powerful, vector-native memory bank for AI agents. Built with the Protocol-Lattice Go Agent Framework, this server offers persistent, searchable, and shareable memory with multiple database backends.

Turn AI conversations into vocabulary learning: capture new words and grammar mistakes from chat into spaced-repetition flashcards, read your review queue and known words, and reinforce vocabulary with SM-2. Remote, OAuth 2.1, EU-hosted.

Mnemex is a Python MCP server that provides AI assistants with human-like memory dynamics through temporal decay and natural spaced repetition, storing memories locally in human-readable JSONL and Markdown formats.

Persistent cross-agent semantic memory for AI agents. Recall past sessions, share knowledge across agents. Multi-agent (isolated writes, shared reads), local-first (SQLite + FTS5), works with any LLM — local Ollama at $0 or cloud APIs like Gemini and OpenAI. Integrations for Claude Code, Claude Desktop, and OpenClaw.