Plugins
ByteRover context engine plugin for OpenClaw (mono build) - in-process recall plus the brv_record tool for durable ByteRover memory writes.
Lossless Context Management plugin for OpenClaw — DAG-based conversation summarization with threshold compaction
Automatic associative memory recall for OpenClaw agents — surfaces relevant context before the model processes each message
智能上下文保护插件：主动管理 tool result 大小、监控 context window 压力、自动触发压缩
Auto-match user delivery task with available skills for better agent responses
Cut OpenClaw token costs. Caveman-speak output mode, context file compression, auto-pruning, savings tracking.
Always-on memory for OpenClaw agents — a context engine that injects a curated MEMORY.md + USER.md into every turn (including group/channel sessions), so bots stop forgetting context and rules.
Gmail watcher for OpenClaw agents via Google Workspace CLI (gws)
RTK token compression plugin for OpenClaw — compresses exec tool outputs to reduce token consumption
OpenClaw plugin that evolves the agent's system prompt using SCOPE (Self-evolving Context Optimization via Prompt Evolution)
MemClaw Context Engine - Native context management for OpenClaw with automatic recall and capture
Graph-backed persistent memory engine for OpenClaw. Replaces the default context window with SurrealDB + vector embeddings that learn across sessions.
OpenClaw plugin that recommends locally loaded SkillHub skills.
Smart session compaction plugin for OpenClaw to manage token usage and support unlimited conversation length.
OpenCode plugin — Agent-lifecycle intelligence layer for OpenTalk2HTML-NotMD. Intent routing, context compression, quality gates, and document state management.
OpenClaw Plugin SDK hook for per-turn context compression via Compresh — episodic memory architecture for LLMs
Keeps long OpenClaw chats focused by trimming old noise, shortening older context, and auto-triggering compaction under pressure.
Reduce LLM token consumption by 60-90% by filtering CLI output through snip
