OmniRoute — управление кешем LLM-ответов

★ 7.9 · general

omni-cache is a Claude Code skill that manages the LLM response cache in OmniRoute — covering cache statistics, entry clearing, TTL policy configuration, and semantic-similarity caching thresholds. It exposes four endpoints: GET and DELETE for both `/api/cache` and `/api/cache/stats`, all secured with a Bearer token or session cookie. The detailed stats endpoint returns per-layer cache metrics, giving visibility into hit rates and overall caching efficiency. Developers and DevOps teams use it to reduce token costs by reusing stored responses for semantically similar or repeated requests.