3 lines of code. Auto-classification. Hybrid retrieval. Multi-entity recall. Connect your data sources and make your AI agents smarter.
from mindglue import MindGlue
mg = MindGlue(api_key="mg_...")
mg.store("user:123", "Prefers dark mode, uses Python")
result = mg.recall("user:123", query="user preferences")
# Returns formatted context ready for system prompt injection
Everything your AI agent needs to remember, reason, and recall — without managing embeddings, graphs, or caching yourself.
Every memory is automatically classified as a fact, preference, episode, or procedure. No manual labeling needed.
Vector similarity + knowledge graph + temporal recency. Three retrieval signals fused into one ranked result.
Recall memories across users, projects, and deals simultaneously. Perfect for complex agent interactions.
Full multi-tenant isolation. Each namespace is a separate memory universe with its own quotas and access control.
Automatic deduplication catches paraphrases at 0.85 similarity. Duplicates are merged via LLM, preserving version history.
Python, TypeScript, and Go SDKs. MCP server for Claude Desktop. Real-time WebSocket streaming. Full REST API.
Automatically sync data from the tools your team already uses. Every record flows through the full intelligence pipeline — classify, extract entities, embed, dedup, store. No custom code needed.
Database rows
Docs, Sheets, PDFs
Pages & databases
Channel messages
CRM records
Tickets & articles
Scale plan includes 3 connectors. Additional connectors $19/mo each. Enterprise: unlimited.
Start free. Scale as you grow. No hidden fees.
Start free. No credit card required. Upgrade when you need more.
Create Free Account