The File-First Memory Runtime for AI Agents
Store decisions as simple markdown files — versioned, portable, and always there when the next session starts.
The Problem
Every new session starts from zero.
You brief your agent on project conventions, domain rules, and architecture. It gets it. Next session — a blank stare. You re-explain everything, only for it to contradict yesterday's decisions.
Local scratchpads only shrink the problem. Every tool, machine, and collaborator ends up with fractured, uncoordinated memory.
Divergent.
Your desktop, remote containers, and teammates' agents each remember conflicting, fragmented context.
Invisible.
Black-box vector databases hide what agents have learned, preventing inspection, verification, or manual correction.
Unreachable.
Serverless functions, CI/CD runners, and hosted workflows start from scratch on every invocation with zero ground truth.
Runtimes
One engine, three storage modes
MemoFS is built on a unified store abstraction. Choose where your memory resides based on your workflow and access needs.
Local Storage Mode
All memory is stored directly on your project's filesystem as versioned markdown and JSON. Zero network latency, fully offline, and 100% git-trackable.
Hybrid Sync Mode
Sub-millisecond local execution with automatic cloud replication. Your agent memory follows you across multiple machines, desktop workstations, and teammates.
Hosted MCP Endpoint
MemoFS Cloud hosts the runtime directly. Remote agents, ephemeral CI/CD containers, and serverless bots connect via standard MCP endpoint URL and API key.
Features & Capabilities
Engineered for autonomous agent memory
From sub-millisecond local hybrid recall to self-healing code anchors and multi-agent coordination. Built with transparent files, deterministic logic, and zero black boxes.
Code Anchoring & Drift Detection
Designed specifically for coding agents: memories bind directly to source code files, byte hashes, and AST symbol paths. When underlying code changes, query-time drift detection automatically demotes stale memory relevance.
Hybrid Recall & Gated Fusion
Sub-millisecond BM25 lexical keyword matching fused with dense vector embeddings via bounded gated-bonus ranking. Unifies memory chunk identity to eliminate duplicate hits and avoid LLM context bloat.
Entity Graph & Consolidation
Extracts entities and relationships into transparent nodes.jsonl and edges.jsonl files. Deterministic and critic-gated consolidation merges redundant facts and automatically retires conflicting assertions.
AgentFS Workspace Scratchpads
Isolated scratch filesystem per agent turn with outcome-gated memory promotion (success, failure, aborted). Automatically preserves and promotes high-signal durable facts while pruning intermediate session noise.
Agent Behavior Enforcement
Deterministic push hooks across 9+ agent environments including Claude Code, Cursor, Copilot, Codex, and Gemini CLI. Injects relevant memory context before tool execution and preserves state across context compactions.
Cognitive Decay & Cold Archive
Memory-kind expiry windows age facts into unverified status without manual maintenance. Operator-guided consolidation moves superseded memories to cold archive with full forensic restoration.
Ephemeral Stream & Agent Pub/Sub
Real-time append-only stream for transient cross-agent coordination and human steering hints. Filtered out of durable recall to eliminate noise, with automated TTL garbage collection on maintenance runs.
Procedural Memory & Playbooks
Actionable step-by-step workflow recipes stored as human-readable Markdown and structured JSON. Context strategist dynamically matches and surfaces relevant playbooks based on agent task type.
Static Memory Linter
Semantic static analyzer CLI for CI/CD pipelines. Detects contradictory assertions across stored notes, identifies dead code anchor references, and flags orphaned knowledge graph entities before merge.
Visual Memory Studio
Local zero-dependency webview debugger for .memofs/. Inspect BM25 vs vector vs graph recall weight breakdowns, explore interactive entity graphs, and test snapshot rollback diffs in real time.
Multi-Language SDKs
Native reader libraries and typed JSON-RPC clients for Python, Rust, and Go runtimes. Built from version-locked Open Memory Schemas and verified against a shared cross-language test suite.
Immutable Action Receipts
Append-only audit receipts capturing consequential agent actions, reversibility metadata, argument digests, and before/after snapshot hashes for complete execution accountability.
Why File-First
MemoFS vs. Black-Box Memory
Most memory tools hide your data in opaque remote dashboards. MemoFS keeps everything transparent, diffable, and versioned in `.memofs/`.
| Dimension | MemoFS | Hosted Memory DBs |
|---|---|---|
| Where memory lives | ✓Plain markdown files in your repo | ✕Locked in a proprietary cloud database |
| Inspect & edit | ✓Any editor (VS Code, Cursor, Vim) | ✕Vendor web UI only |
| Version control | ✓Git-tracked alongside your codebase | ✕Separate external system |
| Data ownership | ✓100% owned and controlled by you | ✕Subject to vendor lock-in & terms |
| Offline execution | ✓Full offline support by default | ✕Requires persistent network connection |
| Protocol compatibility | ✓Native Model Context Protocol (MCP) | ✕Custom proprietary SDKs only |
One command. Your agent never forgets.
Get started in under 2 minutes. Install the CLI or import the SDK directly into your TypeScript project.