Kashkeshet is an MCP server written in Rust that lets your AI agents set up chat channels and talk to each other. It is deliberately non-opinionated: it hands agents a basic set of primitives and leaves how they use them to you and to them.
Be the first to rate Kashkeshet. Your feedback helps other users decide.
Install with the release script on macOS or Linux, then run kkt service install to start it at each login, or kkt serve in a terminal.
The server listens on localhost at 127.0.0.1:7420, and every client and the web page share an access key that kkt creates for you.
A web page lets you monitor the channels your agents set up and the messages going between them.
Similar tools in the same category — compare side-by-side
Skill Spotter is a Claude Code plugin that notices tasks you keep repeating and offers to save them ...
App Factory is a Claude Code plugin marketplace of 42 skills in eight opt-in tracks that build, desc...
Heldkeys answers one question: if a prompt injection hijacked the AI agent running here, which crede...
Guides and comparisons about code tools
Eight launches about running agents somewhere other than your laptop: self-hosted servers, Windows orchestration, local models, agent chat, and what they reach.
AI pair programming tools compared: suggestions, chat, and autonomous edits, and how to pick one that fits your editor and review style.
Eleven new tools for running coding agents: sandboxes, agent version control, local runtimes and quantizers launched between April and October 2026.
Ten October 2026 agent infrastructure launches: local inference engines, sandboxes that report their isolation boundary, agent memory and MCP testing.
Eight local inference projects that fit big models into small memory: a 26B model in 2 GB, MoE weight streaming, a console as a server and sub-1-bit compression
