ProvanceMedia

Yodacode — Communication skill for Claude Code

Communication community

Personal Claude-Code-powered chat agent for Slack, WhatsApp, and beyond.

How to install Yodacode

This entry records only its repository, not the path inside it, so there is no exact command to give. Open ProvanceMedia/yodacode and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Yodacode does

Personal Claude-Code-powered chat agent for Slack, WhatsApp, and beyond.

Alternatives in Communication

Open SaaS — A 100% free modern JS SaaS boilerplate (React, NodeJS, Prisma) 14.1k ★
Slack — Communication Slack integration via HTTP/OAuth MCP 14k ★
Om Slack Scan — Slack Deep Scan 4.6k ★

README

██╗   ██╗ ██████╗ ██████╗  █████╗  ██████╗ ██████╗ ██████╗ ███████╗
╚██╗ ██╔╝██╔═══██╗██╔══██╗██╔══██╗██╔════╝██╔═══██╗██╔══██╗██╔════╝
 ╚████╔╝ ██║   ██║██║  ██║███████║██║     ██║   ██║██║  ██║█████╗
  ╚██╔╝  ██║   ██║██║  ██║██╔══██║██║     ██║   ██║██║  ██║██╔══╝
   ██║   ╚██████╔╝██████╔╝██║  ██║╚██████╗╚██████╔╝██████╔╝███████╗
   ╚═╝    ╚═════╝ ╚═════╝ ╚═╝  ╚═╝ ╚═════╝ ╚═════╝ ╚═════╝ ╚══════╝

Your own AI assistant for Slack, running on your server.

It runs on a subscription you already pay for, **Claude** or **ChatGPT**. No API key, no per-request billing. One command sets it all up, and you're DMing your bot a few minutes later.

What is this?

A personal AI agent that lives on your own server and answers you in Slack. Every reply runs through a real coding agent, **Claude Code** or **ChatGPT Codex**, whichever you prefer. So it has real tools: it can read and write files, search the web, drive a browser, delegate to subagents, and use a full shell.

It runs as two Docker containers. The **broker** holds your API keys. The **agent** is the bot itself, and it never sees them. More on that under [Security](#security-de-rooted-by-default).

It also remembers. Memory lives in plain files, gets tidied up daily, and is fully searchable, so your assistant builds up context over time instead of starting cold every conversation.

What you need

A small always-on Linux server. A cheap VPS, a cloud droplet, a box under the desk. A laptop that sleeps won't work, since the bot needs to stay connected. Give it 1 GB of RAM at the very least, 2 GB to be comfortable. Below that the kernel starts killing replies halfway through, though the installer offers to add swap on small machines.
Docker. If you haven't got it, the installer offers to put it on for you.
A Claude subscription or a ChatGPT one. Either works. The installer asks which and signs you in, and there's no API key involved. docs/ENGINES.md covers what