"Koa" matching MCP servers:

Enables image generation using Gemini native models, supporting both single prompts and batch processing via a file-based queue. It allows for detailed configuration of aspect ratios and models using YAML frontmatter across various MCP-enabled clients.

Blocks prompt injection, jailbreaks, and out-of-scope input before an AI agent acts. classify_input tool, stdio transport.

A lightweight short-term memory MCP server that automatically stores and recalls working context, session state, and task progress for AI agents. Memories auto-expire after 24 hours and integrate seamlessly with workspace-aware storage across multiple projects.

A bridging framework that integrates Knowledge Organization Infrastructure (KOI) with Model Context Protocol (MCP), enabling autonomous agents to exchange personality traits and expose capabilities as standardized tools.

A local gate that lets agents submit risky tool actions for policy evaluation and receive ALLOW, DENY, or REQUIRE_APPROVE decisions before executing irreversible side effects. It can also open human approval holds and record an activity trail without executing the side effect itself.

Enables users to leverage a mixture-of-agents architecture where multiple AI models discuss a question in parallel and a president model synthesizes the best answer.