The Soul of your AI.
A local, portable, and universal 🧠 for your AI
MEMM is an AI-native app that turns your knowledge into a second brain for you and your AI. Stop re-explaining context and burning tokens. Build a permanent memory that captures your reasoning, improves results, and grows over time.
Built on files. Owned by you.
Also available for Windows and Linux · View on GitHub ↗
Better retrieval precision over any vector-only RAG model. Consistent, reliable results.
Stop burning money on noise. Inject only high-signal reasoning that actually matters.
Pure Rust scoring engine. Sub-millisecond contextual awareness at any scale.
Your AI has amnesia.
Context engineering is still manual. Three problems every engineer knows too well.
The blank slate.
Every conversation starts from scratch.
You've explained your error handling conventions a hundred times. ChatGPT, Claude. None of them retained any of it.
The bloated context file.
CLAUDE.md grows until you stop trusting it.
No structure. No scoring. No way to know what the AI actually reads. A 2,000-word file injected wholesale into every query.
The silo tax.
Your ChatGPT has no idea what your Cursor knows.
Every tool has its own memory. You maintain three diverging copies of the same knowledge by hand, forever.
Give your AI a memory that
grows and refines over time.
Your memory, in plain text.
Memories are Markdown files with YAML frontmatter. Open them, edit them, version with Git, share with your team. No database. No embeddings. No black box.
---
id: error-handling-conv
type: concept
importance: 0.9
tags: [rust, errors, conventions]
related: [api-design, rust-patterns]
---
## Error Handling Convention
Always use Result<T, AppError>.
Propagate with `?` operator.
Never panic in library code.Scored, tiered, token-aware.
A 6-signal engine (BM25 · semantic · graph · recency · importance · frequency) ranks every memory per query. L0 / L1 / L2 tiers load exactly the right level of detail. No over-stuffing. No blind spots.
Connect once, works everywhere.
One MCP server exposes your memory to every tool simultaneously. A single ground truth. No duplication. No diverging copies of the same knowledge.
Memory that compounds, not rots.
A built-in governance layer surfaces stale memories, contradictions, redundancy, and debris. A 0–100 health score tracks quality. It gets better with every session, automatically.
Set up your own mini Palantir 🔮
We bring the magic of enterprise ontologies to your personal AI. Like a mini Palantir, but local, open, and yours. Every memory is semantically typed so your AI reasons over structured knowledge, not just flat text.
Entity
Real-world objects, people, APIs, databases, services. The nouns your project talks about.
Concept
Abstract ideas, patterns, conventions, architectures. The principles that guide decisions.
Source
External references, docs, papers, specifications. Protected ground truth you can cite.
Synthesis
Distilled conclusions, decisions, post-mortems. The output of reasoning, ready to reuse.
Your AI doesn't just search. It reasons over a structured knowledge model.
Born for the AI Era.
Legacy tools solve human amnesia. MEMM is built to solve AI amnesia.
A different epoch.
Human-first note-taking adapted through a patchwork of 1000+ plugins. Convoluted, fragile, and high cognitive load. Built for a problem that didn't exist when they were created.
- +1000 plugins to duct-tape AI
- Manual link maintenance
- No context scoring or governance
- Amnesic by default
- Doesn't learn between sessions
Integrated from day one.
A context engine built from line zero for the LLM era. No plugins. A direct, high-fidelity bridge between your knowledge and every AI tool you use.
- Zero-config intelligence
- Automated knowledge scoring
- Built-in governance & health
- Native MCP: ChatGPT, Claude, Cursor
- Grows & compounds with every session
From files to compound intelligence.
Three steps to a memory worth trusting.
Build your workspace.
Create your first memories: who you are, your stack, your conventions, your architectural decisions. Drop external documents into inbox/ and promote them to the right category.
Connect your tools.
Point ChatGPT, Claude, or Cursor at your MCP server. They call get_context with every query and receive the most relevant memories, scored and budgeted automatically.
Let it compound.
Every session adds knowledge. The scoring engine learns what you use. Governance surfaces what needs updating. Your AI gets smarter with your project, permanently.
The Technical Matrix.
Compare the architecture of MEMM OS against traditional vector systems, legacy tools, and manual approaches.
| Capability | MEMM OS | Vector RAG | Manual (CLAUDE.md) | Legacy (Obsidian) |
|---|---|---|---|---|
| You can read the memory | ✓ | ✗ | ✓ | ✓ |
| Inspectable retrieval | Yes (Transparent) | No (Blackbox) | Manual only | No |
| See what AI will retrieve | ✓ (Simulation view) | ✗ | N/A | N/A |
| Ontology | ✓ Native | ✗ Flat vectors | ✗ | ✗ Tags only |
| Real-time token budgeting | Yes (L0/L1/L2) | Static Window | None | None |
| Works across all tools | ✓ (Native MCP) | ✗ Siloed | Partial | ✗ Plugins |
| Local model support | Native / Hybrid | Limited / API | High | Via Plugins |
| No database infrastructure | ✓ Zero infra | ✗ Vector DB | ✓ | Local App |
| Context health & decay | ✓ Automated | Static DB | Rotting | Human only |
| Scales with complexity | ✓ | ✓ | ✗ | Partial |
| Built-in governance | ✓ | ✗ | ✗ | ✗ |
| Compounds over time | ✓ Automatic | ✗ | Manual | Manual |
Built for Engineers.
What's under the hood before you download anything.
Your Personal Knowledge Ontology
Inspired by the same ontology-driven approach that powers enterprise intelligence platforms, but rebuilt for individuals. Every memory is typed as Entity, Concept, Source, or Synthesis, giving your AI a structured world model instead of a bag of words.
Fair Source → Open Source
Source code available on GitHub to download and audit. Built under the FSL license with automatic consolidation to Open Source (Apache 2.0). Human-readable, documented formats. Your memory is yours forever. No vendor lock-in.
Privacy by Design
Nothing leaves your machine. No analytics, no telemetry, no cloud sync. The MCP server binds to 127.0.0.1 only. Your knowledge stays yours, permanently.
Start with what you know. Files.
MEMM is free, open source, and turns your knowledge into a compound second brain. Stop re-explaining and start building with an AI that finally remembers.
Read the academic paper · Read the technical whitepaper · View on GitHub