Unified client SDK for KB Labs platform. Zero dependencies. Access LLM, Cache, VectorStore, Analytics, and any adapter through a single client.

SQLite storage adapter for @v3cord/llm-cache

Redis storage adapter for @v3cord/llm-cache

Semantic caching layer for @v3cord/llm-cache

CLI for @v3cord/llm-cache

Qdrant vector database adapter for llm-cache semantic search

HTTP service wrapper for llm-cache

Redis storage adapter for llm-cache

Observability, metrics, logging, and tracing for llm-cache

DynamoDB storage adapter for llm-cache

Cost tracking and pricing calculations for llm-cache

Core caching engine for llm-cache — semantic and exact-match caching with embedding-based similarity

Content-addressed model response cache for dsh-rerun attempts: replays a byte-identical request, records a new one

First-message LLM provider plugin for DeepSeek Harness session titles

Read-only performance diagnostics for DeepSeek Harness: session load (open/restore) timing, spill-hit counts, compaction count and trigger, context-injection volume (AGENTS.md/skills/tool-schema token share), and LLM cache hit rate — surfaced via the /fas

A cache object that deletes the least-recently-used items.

A lightweight cache for file metadata, ideal for processes that work on a specific set of files and only need to reprocess files that have changed since the last run

Shared anonymous user identity for DeepSeek Harness telemetry and feedback correlation

Require hook for automatic V8 compile cache persistence

A simple key/value storage using files to persist the data