Facthouse
Related Servers
Alternatives to Facthouse
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityDmaintenanceA portable MCP server providing a shared intelligent memory system for any MCP-compatible AI tool, enabling storage, retrieval, extraction, and governance of memories across sessions.16 npmMIT
- FlicenseNot gradedqualityDmaintenanceA persistent, searchable memory layer for AI assistants, exposed as an MCP server. It supports CRUD, namespaces, hybrid ranked search, relationships, chat-context assembly, import/export, and optional LLM-powered auto-extraction.-
- AlicenseNot gradedqualityBmaintenanceOpen-source MCP memory server providing persistent, cross-platform context for AI tools via a knowledge graph with encrypted storage.7 npm13AGPL 3.0
- AlicenseNot gradedqualityDmaintenanceMCP server that enables persistent, hybrid, local memory for LLM agents, with vector + BM25 search, knowledge graph, and policy-driven retention, providing token-budgeted context injection for AI assistants.MIT
- FlicenseNot gradedqualityDmaintenanceAn MCP server for managing persistent AI memory using hybrid search (keyword + semantic vector) with SQLite storage and offline-first local embeddings.-
- AlicenseNot gradedqualityCmaintenanceA self-hosted MCP server that provides a personal semantic memory layer for AI tools. It enables storing, searching, and managing memories using hybrid vector and keyword search, allowing AI assistants to recall information by meaning.MIT
TDQS
Scored across 10 tools
The recall trio (search_knowledge, get_entity, get_context) genuinely overlaps — all three can answer 'what do you know about X' — but the descriptions draw explicit boundaries (fast fact vs. subject dossier vs. relationship traversal) and cross-reference each other. capture_fact vs. log_event is cleanly separated as meaning vs. raw utterance, and the utility tools (get_stats, get_schemas, consolidate, get_session_context) are unambiguous. Only a careless agent would misselect, but the overlap is real enough to keep this below 5.
Consistent snake_case with a strong verb_noun pattern throughout: search_knowledge, get_entity, get_context, get_schemas, get_stats, log_event, get_events, capture_fact, get_session_context. The single bare imperative 'consolidate' is the one deviation. Easily predictable for an agent.
Ten tools is well-scoped for a persistent-memory store: capture, retrieve (three granularities), inspect, log, and batch-consolidate each earn their place. No redundant aliases and nothing obviously missing from the count itself.
Read and write paths are thorough — capture, search, entity/topic traversal, raw event log, stats, schema listing, and consolidation for dedup/contradiction. However, there is no update or delete/forget operation for a stored fact or entity, which is a notable gap for a memory store where users will ask to correct or remove knowledge; the server expects reconciliation to handle it implicitly. Coverage of the lifecycle is therefore write-heavy but not closable by the agent.