InfiniteContext — Tiered Memory for AI

The named artifact (test, doc, code path) exists in the repo today and is runnable. Per-claim, not per-project — a claim can be shipped while the parent project is still in development.

• Exact flat vector store with persisted, rebuildable index artifacts (save/load/merge/split), plus an approximate HNSW backend for retrieval at scale

  src/core/VectorStore.ts (persisted exact flat artifacts), src/utils/IndexManager.ts (flat + hnswlib-node HNSW, graceful flat fallback), src/core/HierarchicalRetriever.ts (per-trace HNSW ANN above a 2,000-item threshold); tests/core/VectorStore.test.ts

• Hierarchical bucket memory with a multi-strategy retriever

  src/core/{Bucket,MemoryManager,HierarchicalRetriever}.ts; tests/core/{Bucket,HierarchicalRetriever}.test.ts

• Automatic prompt/output categorization with keyword, vector-similarity, and adaptive strategies

  src/categorization/PromptCategorizer.ts + strategies/

• Multi-level summarization via OpenAI embeddings/completions

  src/summarization/SummarizationEngine.ts

• Negation-aware retrieval (Widdows orthogonal negation)

  src/core/NegationAwareRetrieval.ts; tests/core/NegationAwareRetrieval.test.ts

• Pluggable storage tiers — local disk and Google Drive behind a common provider interface

  src/providers/{StorageProvider,LocalStorageProvider,GoogleDriveProvider}.ts

• Continuous integration on Node 20 and 22 (build + test)

  .github/workflows CI; Jest suite under tests/