jagdeepsinghdev/pi-prefix-cache-compaction

Pi compaction that reads the conversation's prompt cache, with a cache-expiry timer

Hermes Agent context engine: the main model writes the compaction handoff on its cached prompt prefix

Optimize DeepSeek prefix cache hit rates — reduce API input costs by 95%+ in multi-turn coding sessions

稳定可用 Use billion-context instead .Active Context Pruning for pi ，100K tokens is enough. ，主动压缩上下文，10w 上下文足矣

CliffCompaction — autocompaction for long-horizon coding agents

Improve Pi prompt/KV cache hit rates with stable prompts, OpenAI-compatible cache keys, proxy compat warnings, and footer cache stats.

Use OpenAI Responses API native compaction in OpenCode

Context Management processor for Pydantic AI agents, providing LLM-powered summarization or zero-cost sliding window trimming to handle infinite/long-running conversations without context overflow. Supports flexible triggers, safe cutoffs, and custom prompts for efficient AI apps.