$ loading_
Optimize LLM context windows and rotate sessions in long agent workflows.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "@yavdaanalytics/context-optimiser" yet — see the docs or source repo.
Compact the existing chat history in the current agent workflow. Keep goals, key constraints, completed steps, and pending tasks while reducing context usage.
A condensed context summary that preserves the key information needed to continue the task.
The current session is nearing the context window limit. Create a new session and carry over the essential state required to continue the task.
A rotated session handoff that lets the agent continue the original task in a new session.
Organize the key state of the current multi-step agent workflow, remove redundant content, and output a compact context for subsequent calls.
A deduplicated task-state summary suitable for continued model calls.
Developers building multi-turn, multi-step AI agent workflows can use it to compact chat history and avoid interruptions caused by oversized context. This helps tasks keep running across longer flows.
When the current session is close to the context window limit, it can rotate to a new session while preserving the core information needed to continue. This fits scenarios that require extended, continuous reasoning.
It is an MCP server for context window optimization and session rotation in agentic workflows. Its goal is to prevent LLMs from running out of context by compacting chat history and rotating sessions.
It is best suited for long conversations and multi-step agent tasks. It is especially useful when model context windows are likely to be filled by accumulated message history.
The provided information only indicates that it is an MCP tool and does not include installation steps, runtime details, or key requirements. See the source repository for specifics.
Monitor AI coding context usage and preserve state before compaction.
Compress context and persist checkpoints to cut AI agent token usage.
Manage contextual data in Markdown with metadata for save, search, and retrieval.
Provide persistent virtual memory, retrieval, and prompt assembly for AI context.
Give AI coding agents persistent local memory across sessions.
Manage AI project context, decisions, and constraints across sessions with Git.