New Open-Source Python Tool Gives AI Coding Agents Human-Like Memory With Zero Dependencies

A new open-source Python tool called 'mind' brings human-like memory to AI coding agents using spreading-activation recall and Ebbinghaus forgetting curves — all in a single, zero-dependency file — achieving perfect recall benchmarks with sub-14ms latency across graphs of up to 1,000 nodes.

• A new open-source tool called 'mind' provides brain-like memory for coding agents using spreading-activation recall, Ebbinghaus forgetting curves, and deterministic dream consolidation — all in a single Python file with zero dependencies and no API keys required.
• The system operates across three layers mimicking human memory: a working memory layer injected into agent rule files, a hippocampus layer using a weighted concept graph with fuzzy recall and IDF ranking, and a cortex layer for consolidated long-term knowledge promoted during automated dream cycles.
• Benchmarks show perfect recall@1 and recall@5 on graphs up to 1,000 nodes with sub-14ms latency, full multilingual support across 10 languages, and a 267-test suite with adversarial fuzzing and mutation testing — while explicitly noting limitations in cross-domain synonymy, morphological analysis, and scalability beyond roughly 10,000 nodes.