mcp-server

ShibaClaw: Secure Self‑Hosted AI Agent for Founder Automation

Name: ShibaClaw: Secure Self‑Hosted AI Agent for Founder Automation
Author: RikyZ90

Deploy a security‑first AI assistant with multi‑provider support, chat integrations, and scheduling—all locally hosted for startups that need control and flexibility.

From the repo · “🐕 Self-hosted security-first AI agent · 28 providers · 11 chat channels · WebUI · 3-level memory · task-schedule · automation · skills · MCP”

intermediate⏱ 1-2 hours💵 Free (self-hosted)

84 stars+2★ this week10 forksPythonGuide quality 9/10 · reviewed Oct 22Updated 10/20/2026100% free · open source

What it is

Create a custom AI agent from scratch

Real use case example

"A founder wants to automate the onboarding process for new customers by adding them to a Notion database and sending a welcome email. To set this up, they would install a self-hosted AI agent (like ShibaClaw), connect their Stripe webhook, and define a task to automate the Notion database updates and email sending. The outcome: automated onboarding for their customers without the need for manual intervention."

Best insideClaude Desktop

When to use it

•When you need a customized AI solution for your startup's specific security and automation requirements
•When you want to integrate AI capabilities with your existing infrastructure and tools
•When you require a high degree of control and flexibility in your AI agent's configuration and behavior

Ready-to-paste prompt

Add a new provider by running `python add_provider.py --name MyProvider --type MCP` and then configure it in the `config.json` file

Heads up: Make sure you have Python 3.8 or later installed, as ShibaClaw is built with Python and requires this version or higher to run

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/shibaclaw && curl -fsSL https://workflowstacks.com/api/skills/shibaclaw/claude-skill -o ~/.claude/skills/shibaclaw/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How ShibaClaw: Secure Self‑Hosted AI Agent for Founder Automation works

Codeflow

Free to inspect

ShibaClaw: Secure Self‑Hosted AI Agent for Founder Automation is a very large Python project (~71k lines across 268 code files, plus 121 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, Apache-2.0 license, has a test suite.

Size

Very large codebase

~71k lines · 268 code files · days to read — use, don't read

Setup

Developer setup

A real software project. Use it via its install path; don't expect to read it all.

Runs on

Python · Docker

No API keys detected

Python 63%JavaScript 23%CSS 6%HTML 6%

Where to start reading

1
README.md
Start here — what it does and how to install it
2
shibaclaw/__main__.py
Where the program starts running
3
pyproject.toml
Dependencies and the commands it exposes
4
shibaclaw/__init__.py
Inside shibaclaw/ — the main logic begins here

What's in each folder

shibaclaw/Core code — the actual logic327 files

plugins/Sub-packages / plugins12 files

docs/Documentation7 files

scripts/Helper scripts9 files

pyinstaller-hooks/Folder3 files

tests/Tests — proof it works110 files

assets/Images & static assets12 files

.github/CI / automation (GitHub Actions)2 files

READMEHas testsDocumentedCI checksDocker readyApache-2.0 licenseUpdated this month

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/shibaclaw && curl -fsSL https://workflowstacks.com/api/skills/shibaclaw/claude-skill -o ~/.claude/skills/shibaclaw/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

Quick Actions

Details

Creator

RikyZ90

Language

Python

Do this with a working workflow

Pages we wrote and tested that put a skill like this to work — importable n8n templates, outcome guides and Claude Desktop configs.

Free n8n template

AI Support Agent (Chat)

A chat agent that answers customer questions from YOUR business info — with conversation memory. It never invents policies; anything it doesn't know, it hands to a human.

MCP config