A bridge that connects AI agent runtimes to the XMTP open messaging network for secure, decentralized communication.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install xmtp-agents
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install xmtp-agents using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is XMTP Agent Connectivity Skill?
The XMTP Agent skill enables developers to link their AI agents—including OpenClaw and Claude Code—directly to the XMTP decentralized messaging protocol. By implementing a bridge pattern, this skill allows an agent to receive DMs, process them using its full suite of tools and memory, and respond autonomously within a secure environment. This integration is essential for making agents reachable over a permissionless network while maintaining strict control over resource access.
This capability is a cornerstone of the Openclaw Skills ecosystem, providing a robust transport layer for agentic workflows. The bridge distinguishes between the owner and public users, ensuring that sensitive tools and file access are only available to authorized identities while providing a safe, restricted interface for the general public.
XMTP Agent Connectivity Skill Use Cases
- Making an AI agent reachable via decentralized DMs for remote command and control.
- Setting up a secure support bot that interacts with users over the XMTP network.
- Bridging local agent processes to a mobile-friendly messaging interface.
- Creating a gatekept interaction layer where the owner has full tool access while public users have restricted conversational access.
How XMTP Agent Connectivity Skill Works
- Initialize the XMTP identity using the CLI to generate a wallet and encryption keys.
- Resolve the owner's Inbox ID to establish administrative permissions and security guardrails.
- Launch a bridge script that continuously streams incoming messages from the XMTP network using the stream-all-messages command.
- Route incoming text to the agent backend (like OpenClaw) based on the sender's identity to determine the permission level.
- Capture the agent's output and relay it back to the specific XMTP conversation ID to maintain a two-way dialogue.
XMTP Agent Connectivity Skill Setup
Install the XMTP CLI and initialize your environment:
npm install -g @xmtp/cli
xmtp init --env production
Verify your registration and retrieve your inbox ID:
xmtp client info --json --log-level off --env production
Before starting the bridge, resolve the owner's inbox ID using their Ethereum address to ensure the security of your Openclaw Skills deployment:
export OWNER_INBOX_ID=$(xmtp client inbox-id -i "0xOWNER_WALLET_ADDRESS" --json --log-level off --env production | jq -r '.inboxId')
XMTP Agent Connectivity Skill Data Schema & Taxonomy
The skill processes XMTP message objects in a structured JSON format to manage sessions and identities.
| Field | Description |
|---|---|
id |
Unique identifier for the specific message. |
conversationId |
Unique ID for the chat session, used as the agent session ID for persistent context. |
senderInboxId |
The unique identity of the sender, used for permission branching between owner and public. |
content |
The raw text message passed to the agent backend. |
sentAt |
Timestamp of the message for chronological processing. |
XMTP Agent Connectivity Skill Advanced Features
- Multi-backend support: Compatible with OpenClaw, Claude Code, and custom Python scripts via stdin/stdout.
- Permission Branching: Automatically distinguishes between the owner (full tool access) and public users (restricted prompt mode).
- Persistent Context: Uses XMTP conversation IDs as session keys to maintain long-term memory across different users and groups.
- Hot-swappable Prompts: Update the public-prompt.md file to change public agent behavior instantly without restarting the service.
- Deterministic Sessioning: Leverages namespace UUIDs for isolated session management in advanced agent runtimes.
SKILL.md
Loading
Related Openclaw Skills
Clawdbot Logs & Diagnostics
A specialized toolkit for monitoring, debugging, and analyzing the performance and costs of Clawdbot AI agent sessions.
System Integrity and Backup
A professional-grade security suite providing encrypted backups, integrity verification, and Greek legal retention enforcement.
System Health Check
A comprehensive diagnostic tool that validates file integrity, directory structures, and security permissions for your Openclaw Skills environment.
OpenClaw Greek Accounting Meta-Skill
A comprehensive orchestration layer that routes business commands across 18 specialized skills to automate the full Greek accounting lifecycle.
cloudphone
A specialized automation skill for executing AutoJS scripts and Android UI tasks on remote cloud devices using the mcporter CLI.
ClickUp Integration
A comprehensive tool for programmatically managing ClickUp tasks and workspaces via Openclaw Skills.
Featured*
