React Star Rate Component

Enhanced DSH composer stats strip: official CNY cost, coherent live agent-team settlement, live LLM/tool timers, provider balance

Minimal OpenCode TUI sidebar for cache hit, estimated tokens, and generation speed.

Input trigger pipeline: '/' and '@' detection, candidate menu, pick routing to registered sources

Node.js atomic and non-atomic counters, rate limiting tools, protection from DoS and brute-force attacks at scale

LRU and FIFO caches for Client or Server

Require hook for automatic V8 compile cache persistence

Replaces pi's footer with a labelled two-row status line: context pressure as a fixed-size meter, cache and cost, the model and effort level, and the latest turn's TTFT and decode throughput in tokens/second. An opt-in panel (ctrl+q or /breakdown) breaks

Multi-tier exact-match cache for AI agent workloads backed by Valkey. LLM responses, tool results, and session state with built-in OpenTelemetry and Prometheus instrumentation.

A Redis store for the `express-rate-limit` middleware

Parses Cache-Control and other headers. Helps building correct HTTP caches and proxies

Production caching seam for AI workloads (keyless/offline memory reference + Redis adapter): exact-match cache, verified semantic cache, embedding cache, singleflight request coalescing, negative cache, and epoch-based invalidation. Makes the gateway + RA

Reads and caches the entire contents of a file until it is modified

Live prompt-cache visibility in the Pi footer — hit/miss rates, token usage, and estimated cost savings

TypeScript definitions for http-cache-semantics

DeepSeek prefix-cache extension for pi — hit-rate telemetry, prefix guard, and cache-friendly compaction.

Require extension that allows for caching/precompiling

Yet another auto expiring, max sizable cache

React Star Rate Component

Basic object cache with `get`, `set`, `del`, and `has` methods for node.js/javascript projects.