Main

To keep memory helpful (not noisy):

• Start with local SQLite memory to validate what your agents store, then move to a shared cluster once the schema and access policies are clear.
• Use contradiction flags as prompts for clarification, not as “truth adjudication”.
• Periodically review what gets remembered: prefer stable preferences, decisions, and constraints over raw transcripts.

README excerpt (verbatim)

YantrikDB

⚠ Correction notice (2026-04-19): Phase 3 benchmark writeups committed earlier today (docs/phase3a/, docs/phase3b/, docs/phase3c/, docs/phase3d/) used a Python simulator for the "structured memory" condition — NOT the actual yantrikdb engine. Rerun with real yantrikdb is in docs/phase3e/. Full audit trail in CORRECTIONS.md. Full corrected findings post coming 2026-04-20.

A memory database that forgets, consolidates, and detects contradictions.

Vector databases store memories. They don't manage them. After 10,000 memories, recall quality degrades because there's no consolidation, no forgetting, no conflict resolution. Your AI agent just gets noisier.

YantrikDB is different. It's a cognitive memory engine — embed it, run it as a server, or connect via MCP. It thinks about what it stores.

Shortest path to try it: MCP setup for Claude Code / Cursor / Windsurf → (one pip install, one config block).

The bigger picture: YantrikDB is the memory layer being built on the road to YantrikOS — an AI-native operating system where agents are first-class primitives, not apps on top. Memory was the bottleneck, so we're shipping it first.

📄 New paper (May 2026): Skill as Memory, Not Document — measuring three failure modes when agent skill catalogs scale to 5,000 skills. Code + raw CSVs reproducible in benchmarks/skill_recall/.

99.9% token savings vs file-based memory

FAQ

Q: Is the MCP server separate? A: README points to yantrikdb-mcp as the recommended starting point for Claude Code/Cursor/Windsurf.

Q: Where is memory stored by default? A: README says it uses a local SQLite file under ~/.yantrikdb/memory.db.

Q: How do I share memory? A: README shows configuring YANTRIKDB_SERVER_URL and YANTRIKDB_TOKEN for cluster mode.