A Model Context Protocol (MCP) server that implements a knowledge graph-based memory system with AI-powered entity and relation extraction
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"iflow-mcp-belindamo-kg-mem","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# KG Memory: Knowledge Graph Memory with a MCP Server A Model Context Protocol (MCP) server that implements a **heavily typed** knowledge graph-based memory system with AI-powered entity and relation extraction. Unlike other memory systems that rely on basic summarization or unstructured storage, our system uses **structured ontologies** to extract meaningful, typed relationships from unstructured text. ## Why Choose a Heavily Typed Approach? Most AI memory systems treat information as unstructured blobs or simple key-value pairs. This system is different - we use **strongly typed ontologies** that define exactly what kinds of entities and relationships exist in your domain. This means: - **Predictable structure** - You know exactly what types of information will be extracted - **Domain-specific accuracy** - Tailored extraction for your specific use case - **Relationship validation** - Only valid connections between entity types are allowed - **Query precision** - Ask complex questions about specific relationship patterns ## Comparison with Other Memory Systems We're the only system that combines **heavy typing** with **local storage** and **native MCP integration**, making us ideal for developers who want predictable, domain-specific memory extraction without external dependencies. | Feature | **kg\_mcp** (Ours) | Graphiti/Zep | Mem0 | Letta | GraphRAG | | ---------------------- | -------------------------------------- | -------------------------------------------------------------- | ---------------------------------- | ----------------------------- | ---------------------- | | **Type System** | ✅ **Heavily typed ontologies** | ⚡ Temporal knowledge graphs | ❌ Unstructured memories | ❌ Unstructured storage | ❌ Community summaries | | **Str…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.