"sage active" matching MCP servers:

Reduces Claude's context window costs by automatically summarizing inactive files to their public interfaces using AST parsing, keeping only the full contents of the currently active file.

Persistent, consensus-validated institutional memory for AI agents. Gives LLMs real memory that survives across sessions - validated through BFT consensus, not just dumped to a file.

A unified MCP server for AWS that enables natural language infrastructure management, cross-service resource discovery, and dependency mapping. It features 30 intelligent tools for cost optimization, incident investigation, and multi-account operations protected by a robust safety system.

An extension that allows Large Language Models to interact with and analyze Active Directory environments through natural language queries instead of manual Cypher queries.

A Model Context Protocol server that enables AI assistants to interact with the Deriv trading API, providing access to active trading symbols and account balance information.

Retired MCP server for byImprint's historical creator-intelligence experiment. It is no longer active.

Enables selective context handoff for Codex by curating past conversation blocks based on a next topic. It preserves essential blocks, removes unnecessary ones, and keeps sensitive data local without modifying the active chat history.

Enables agents to dynamically generate typed MCP tool schemas, validate Python execution ASTs against unsafe imports, and hot-plug sandboxed callable handlers into an active runtime. This allows new tools to be synthesized and safely executed on demand.

Enables orchestrating Jira-driven software workflows by watching tickets, running herdr agents on active issues, and pushing updates to those agents over MCP.

Enables agents to issue and independently verify signed cross-agent handoff attestations, ensuring delegation scope only narrows, lifetimes stay contained, nonces resist replay, and chains stay continuous. Exposes tools for issuing handoffs, verifying them against constraints and ancestor linkage, checking scope containment, and describing the active key adapter.

Provides local, read-only recommendations for reasoning effort, retry or escalation guidance, and integration boundaries, enabling compatible desktop apps to suggest effort levels without modifying the active conversation.

Enables peer discovery and direct messaging between multiple Codex sessions running on a single machine. It allows AI sessions to coordinate, find other active peers by repository or context, and exchange messages via a local broker.

Enables Claude Desktop and MCP-compatible agents to formulate, solve, and certify mathematical optimization problems using production-grade open-source solvers, providing mathematically grounded decisions.

Digital drugs that modify AI behavior through prompt injection, enabling agents to list, take, and manage active drugs across sessions.

Coordinates up to 64 parallel expert agents for complex multi-agent orchestration and intelligent task routing within Claude Code. It enables users to analyze requests, manage hierarchical execution plans, and monitor active orchestration sessions.

Enables active deception for AI agents by injecting decoy honeypot tools that detect indirect prompt injection attempts, instantly terminate hostile sessions, and seal cryptographic incident evidence.

Enables graph analysis and network-flow computations through the Model Context Protocol, including shortest-path search, topological dependency resolution, maximum flow, minimum spanning trees, and PageRank centrality scoring for interconnected agent networks.

A local MCP server that enables an agent to read personal WhatsApp messages, prepare replies, and send them only after user confirmation.

Enables deterministic, step-by-step execution of YAML-defined skills and workflows for AI agents, with progressive disclosure of active-step instructions and tools. It manages workflow state, conditional DAG transitions, loops, and persistent sessions via start_workflow, next_step, and end_workflow.

Enables an active learning tutor that runs in MCP clients like Codex or Claude Code, letting the AI create and manage lessons, read student attempts, and give feedback while the student practices in a local browser panel.