$ loading_
Monitor context pressure, compress history, and restore rules and files across sessions.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "compaction-mcp" yet — see the docs or source repo.
Check the current session's context pressure, compact history when near the limit, preserve confirmed requirements, unfinished tasks, and key code decisions, then produce a resume-ready summary.
Returns a context pressure assessment, a compacted session summary, and the key items to continue development.
Using previously persisted rules and summaries, rehydrate the important file context for this project and explain the current architecture, conventions, and next task to handle.
Outputs the restored project context, including key files, team conventions, and a recommended next-step task list.
Persist the development rules confirmed in this session, including code style, testing requirements, commit conventions, and prohibitions, so future sessions can reuse them automatically.
Generates and saves a structured rule summary that future agent sessions can inherit directly.
Monitor AI coding context usage and preserve state before compaction.
Manage Claude Code sessions with hot restart and context compaction commands.
Compresses LLM conversation context while preserving meaning and reducing token usage.
Compress content through MCP to reduce context size in AI workflows.
Search your entire codebase and feed rich context to coding agents.
Coordinate multiple AI coding agents to build, review, and remember together.