omni-cache

Manage LLM response cache statistics, TTL policies, and semantic-similarity thresholds via API.

Updated Jun 6, 2026

Search Tags:#caching #api #LLM #cache #statistics #semantic #TTL

Ready to run

npx skills add https://github.com/DevMHOne/nextroute --skill omni-cache

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: omni-cache
Source: https://github.com/DevMHOne/nextroute/tree/main/skills/omni-cache
Command: npx skills add https://github.com/DevMHOne/nextroute --skill omni-cache

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manage the LLM response cache. View cache statistics, clear entries, configure TTL policies, and control semantic-similarity caching thresholds.

Core Features & Use Cases

View cache statistics and health metrics for all layers (in-memory, disk, etc.)
Clear specific entries or entire cache to recover space or reset state
Configure TTL policies and semantic-similarity thresholds to optimize cache hit rate

Quick Start

Authenticate and issue a GET /api/cache to view statistics, then optionally clear caches with DELETE /api/cache.

Frequently Asked Questions about omni-cache

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I view LLM cache statistics and health metrics?▼

Clear specific LLM cache entries or entire caches by issuing a DELETE request to the /api/cache endpoint. This allows you to recover storage space or completely reset the cache state across multiple sessions when needed.

How do I configure TTL policies for LLM response caching?▼

Configure TTL policies for LLM response caching by adjusting settings through the consistent API. This lets you control how long cached responses are stored before expiration, optimizing your overall cache hit rate and data freshness.

What is semantic-similarity caching and how do I adjust its thresholds?▼

Semantic-similarity caching matches LLM responses by meaning rather than exact text. You adjust similarity thresholds via the API to control how closely a new prompt must match a cached prompt to trigger a cache hit.

Can I manage cache settings across multiple LLM sessions?▼

Yes, you can manage cache settings across multiple LLM sessions. The API supports viewing statistics, pruning entries, and adjusting TTL and semantic-similarity thresholds consistently across all active and stored sessions.

When should I clear my LLM response cache?▼

Clear your LLM response cache when you need to recover storage space or reset the application state. Using the DELETE /api/cache endpoint removes stale or unwanted entries, ensuring subsequent LLM responses are fetched fresh.