"Context compression techniques and methods" matching MCP servers:

An MCP server that keeps raw tool output out of the context window by sandboxing and compressing results (about 98% reduction), persisting file edits, git operations, tasks and decisions in SQLite with BM25 retrieval so the agent resumes after compaction, and enforcing routing across 17 platforms via MCP and hooks. It also encourages the model to write scripts that compute answers and log only results instead of reading large data into context.

A context window optimizer and session rotator MCP server for agentic workflows that prevents LLMs from running out of context by compacting chat history and rotating sessions.

Compresses multi-turn conversation history by stripping verbose boilerplate and filler tokens while preserving critical tool parameters and named entities, deterministically cutting inference costs by 40-60%. Runs as a zero-dependency Python MCP server or importable module, compatible with Claude Desktop, Cursor, and other MCP runtimes.

Enables AI agents to browse, trade, and create prediction markets on Context Markets. Supports wallet management, order placement, and market creation through MCP tools.

A local MCP server that acts as a policy-based firewall for personal context, letting users control agent access to their data with deny-by-default rules, per-field grants, and an audit trail.

Enables async-first working practices by providing tools to draft decision docs, convert meetings to async artifacts, score status updates, and triage sync vs async tasks. It also offers reference tools for the Open and Async book's principles and coaching prompts.

A Model Context Protocol (MCP) server designed to easily dump your codebase context into Large Language Models (LLMs).

Enables AI agents to compose MCP tool calls into Unix-style pipelines, executing complex workflows server-side and returning only the final result.

Graph RAG based context management engine for coding agents

Enables semantic search of project documentation using hybrid vector and full-text search with fast and deep query modes for immediate results or complex multi-round synthesis.

Enables AI agents to perform corpus analysis tasks such as discovery, search, and data access via the Model Context Protocol.

Indexes your codebase so AI coding agents can search instead of re-reading files, saving up to 94% of tokens.

Enables querying context about the MAM product components and architecture through MCP, allowing agents to retrieve targeted product information on demand instead of loading full documentation.

MCP server providing persistent memory for LLMs via semantic search over captures, structured facts with temporal tracking, and document ingestion. Enables any MCP-compatible client to maintain a searchable knowledge base across sessions.

This MCP wraps Bitrefill public API to allow agents to search for products and shop using cryptocurrencies like Bitcoin, Ethereum, Solana, and many more.

Credit and payments for AI agent developers — over MCP. No crypto required

Provides Stackpress framework context and documentation to AI utilities like Claude Desktop and Cline. Enables AI assistants to understand and work with Stackpress projects by fetching and serving relevant framework information.

Enables AI assistants to save, load, and search conversation context with AI-powered summarization and auto-tagging. Demonstrates semantic intent patterns and hexagonal architecture for maintainable AI-assisted development.

Enables inspection and governance of enterprise ingested context by classifying security clearance tiers, sanitizing cross-department prompts, and enforcing role-based access boundaries to prevent data leaks. It runs as a zero-dependency MCP server compatible with Claude Desktop, Cursor, and other agent frameworks.

Reference MCP server for OCL, an open standard giving AI agents structured, scoped business meaning for Odoo data. Serves published example context packs via get_context, resolve_noun, and explain_field (stdio, Python, Apache-2.0).