yantrikos/yantrikdb
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.
What's novel
Not another vector database. YantrikDB models how memory actually works — temporal decay, semantic consolidation, contradiction detection, knowledge graph, and proactive triggers.
Code Analysis
8 files read · 3 roundsA hybrid Rust/Python cognitive memory engine that combines vector search with a knowledge graph to provide persistent, self-consolidating, and conflict-aware memory for AI agents.
Strengths
Exceptional engineering depth in concurrency safety (lock ordering, read pools, atomic state swaps) and sophisticated cognitive algorithms like entity-aware consolidation and multi-strategy contradiction detection. The schema evolution history (v1-v37+) demonstrates robust long-term maintainability.
Weaknesses
The codebase is extremely dense with a massive public API surface, which increases cognitive load for contributors. While unit tests exist for core logic, the complexity of the 'cognitive' features makes comprehensive integration testing challenging and likely under-tested relative to the feature count.
Score Breakdown
Signal breakdown
Innovation
Craft
Traction
Scope
Evidence
Commits
149
Contributors
1
Files
249
Active weeks
8
Repository
Language
Rust
Stars
20
Forks
7
License
AGPL-3.0