ritual-agent-deployment

Name: ritual-agent-deployment
Author: zunmax

Verified

Deploy a recurring, self-funding sovereign AI agent on Ritual testnet with one command.

74stars

49forks

PowerShell

Installation

# Add to your Claude Code skills
git clone https://github.com/zunmax/ritual-agent-deployment

Getting Started

Guides for using ai agents skills like ritual-agent-deployment.

Caveman: Cut Claude Token Use by 65%
How agent-side prompt compression works, when to use it, and when not to.
What is an AI Skills Marketplace?
Definitions, how marketplaces work, and how to choose between them in 2026.
Getting Started with AI Skills

Security ReportVerified

Last scanned: 6/28/2026

{
  "issues": [],
  "status": "PASSED",
  "scannedAt": "2026-06-28T07:52:48.855Z",
  "npmAuditRan": true,
  "pipAuditRan": true,
  "promptInjectionRan": true
}

README.md

Frequently Asked Questions

What is ritual-agent-deployment?

ritual-agent-deployment is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by zunmax. Deploy a recurring, self-funding sovereign AI agent on Ritual testnet with one command. It has 74 GitHub stars.

Is ritual-agent-deployment safe to use?

Yes. ritual-agent-deployment passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.

How do I install ritual-agent-deployment?

Clone the repository with "git clone https://github.com/zunmax/ritual-agent-deployment" and add it to your Claude Code skills directory (see the Installation section above).

What programming language is ritual-agent-deployment written in?

ritual-agent-deployment is primarily written in PowerShell. It is open-source under zunmax on GitHub, so you can review or fork the full source.

Are there alternatives to ritual-agent-deployment?

Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh ritual-agent-deployment against similar tools.

Agentic AI for Beginners

Build your first AI agent from scratch - tool use, ReAct pattern, memory, deployment

41 minBeginner

Comments (0)

to leave a comment.

No comments yet. Be the first to share your thoughts!

Related Skills

superpowers

by obra

An agentic skills framework & software development methodology that works.

234,966

pi-dynamic-workflows openclaw-voice-call-realtime

Ritual Sovereign Agent

Deploy a recurring, sovereign AI agent on Ritual testnet with one command. No API keys.

A sovereign agent is a smart contract that wakes itself on a schedule. On every wake it runs an AI agent inside a secure enclave (TEE), pays for that run from its own on-chain wallet, and keeps going until the money runs out. It lives entirely on-chain.

You need three things. The script installs everything else for you (foundry, uv, and - on Linux/WSL - curl).

git - to download the code. Most systems already have it; if git --version fails, install it (see below).
A wallet on Ritual testnet with a little RITUAL in it. Create one in MetaMask or Rabby, then get free testnet RITUAL from the faucet: https://faucet.ritualfoundation.org
That wallet's private key (use a throwaway testnet wallet - never a real one). You paste it once, when the script asks.

Installing git (skip if git --version already works):

Windows: install Git for Windows - it also bundles Git Bash and curl - or run winget install Git.Git.
macOS: xcode-select --install (or brew install git).
Linux / WSL: sudo apt install git (Debian/Ubuntu), or your distro's package manager.

Step 1 - Get the code

git clone https://github.com/zunmax/ritual-agent-deployment.git
cd ritual-agent-deployment

Step 2 - Configure

cp .env.example .env

There is nothing you must edit - the defaults work. PROMPT is the task your agent runs on every wake, so change it to anything you like.

Step 3 - Deploy

On Windows, using PowerShell 7 (pwsh):

pwsh -ExecutionPolicy Bypass -File run.ps1

On Windows, using Windows PowerShell 5.1 (powershell, which is preinstalled on every Windows):

powershell -ExecutionPolicy Bypass -File run.ps1

-ExecutionPolicy Bypass lets the script run without changing any system setting (Windows blocks unsigned scripts by default); it applies only to this one launch.

On Linux / macOS / Git Bash / WSL:

bash run.sh

Each agent is its own contract at a fixed address (your wallet plus a salt). On Windows, use pwsh -ExecutionPolicy Bypass -File run.ps1 in place of bash run.sh in the examples below.

bash run.sh                          # deploy your first agent using .env 
bash run.sh view                     # list all your agents
bash run.sh view 0xSomeWallet...     # list the agents of any wallet
bash run.sh topup 0xAgent... 2       # deposit 2 RITUAL into agent wallet
bash run.sh topup 0xAgent...         # deposit the .env DEPOSIT amount

Before funding anything, check the agent's status on the explorer: https://ritual-testnet-explorer.vercel.app/

Read the status, then act on it:

Agent is dead -> do not deposit. The deposit transaction still succeeds on-chain, but the RITUAL is stranded: the agent has no scheduled wake left (the Scheduler reverts with CallNotFound()), so the funds are never spent and cannot be recovered. A dead agent cannot be revived by funding it - deploy a fresh agent instead.
Agent is live but low on RITUAL -> top it up immediately with topup. A live agent keeps itself going by scheduling its next wake on every wake, but only while it has the balance to pay for it. If it drains and misses a wake, the call chain breaks and funding can no longer restart it.

Known limitation: the restart, start, withdraw, and stop functions are not callable - an implementation bug in Ritual's proxy contract makes them revert. The Ritual team must upgrade the proxy contract before these work. Until then there is no way to restart a dead agent or withdraw stranded RITUAL.

.env holds no secrets - only your public address and run settings.

Variable	Purpose
`RPC_URL`	Ritual testnet RPC endpoint.
`CHAIN_ID`	`1979` (Ritual testnet).
`DEPOSIT`	RITUAL to lock into the agent's wallet on deploy. Use a plain number, where `1` means 1 RITUAL and decimals like `0.5` are fine. One wake costs about 0.5 to 1 RITUAL. `1` is the minimum a deploy accepts, and `5` gives about 5 wakes of headroom.
`CLI_TYPE`	Harness type. `6` = ZeroClaw.
`MODEL`	Model id routed through Ritual's gateway (no external key). Default `zai-org/GLM-4.7-FP8`.
`PROMPT`	The task the agent runs on each wake.
`SALT`	Any unique string - changes the agent address. Use a new one per agent.
`LOCK_BLOCKS`	Optional. Blocks a deposit stays locked. Defaults to `100000`.
`KEYSTORE_ACCOUNT`	Written automatically on first run - the name of your keystore.
`WALLET_ADDRESS`	Written automatically on first run - your public address.

Your private key is never written to .env; it lives encrypted in ~/.foundry/keystores. Still, use a testnet burner wallet - do not import one with real funds.

Network	Chain ID	RPC	Faucet
Ritual testnet	`1979`	`https://rpc.ritualfoundation.org`	https://faucet.ritualfoundation.org

This tool signs transactions with a key it stores in an encrypted keystore under ~/.foundry/keystores - use a testnet burner wallet, never one with real funds. The deposit you lock funds the agent's scheduled runs and is spent over time - it is not recoverable on a whim. This is testnet software, provided as-is, without warranty, and has not been audited. Use at your own risk.

Released under the MIT License. Built by Zun.