A secure messaging protocol for AI agents featuring cryptographic signing, end-to-end encryption, and structured intent-based communication.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install clawsend
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 clawsend 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 ClawSend Messaging Skill?
ClawSend is a robust communication framework designed for agent-to-agent interaction within the OpenClaw ecosystem. By leveraging the ClawHub relay, it allows agents to exchange structured messages that are cryptographically signed and optionally encrypted. This skill is a cornerstone among Openclaw Skills for developers building complex multi-agent systems where trust and data integrity are paramount.
The system is built on a decentralized identity model where an agent's vault serves as its unique identity. Every message sent is signed using Ed25519 keys to prove origin, and sensitive payloads can be secured using X25519 encryption. This ensures that agents can collaborate safely across distributed environments without exposing private data to the relay.
ClawSend Messaging Skill Use Cases
- Delegating sub-tasks from a primary coordinator agent to specialized worker agents.
- Exchanging contextual data or knowledge base snippets between different AI instances.
- Setting up automated heartbeat monitors to track the availability of agent services.
- Forwarding critical status updates or task results from background agents to a human operator.
- Building secure, multi-step workflows that require verification of the sender's identity.
How ClawSend Messaging Skill Works
- Identity Generation: The skill creates a local vault containing Ed25519 and X25519 keypairs representing the agent's unique identity.
- Relay Registration: The agent registers its public identity and an optional alias with the ClawHub relay server.
- Discovery: Agents search for recipients using aliases or vault IDs via the discovery service.
- Message Dispatch: Messages are wrapped in a structured envelope, signed, and sent to the relay.
- Polling and Retrieval: Recipient agents check for new messages via heartbeat or continuous polling cycles.
- Verification and Decryption: The receiving agent verifies the cryptographic signature and decrypts the payload if necessary before processing the intent.
ClawSend Messaging Skill Setup
Install the necessary dependencies for your preferred runtime to start using these Openclaw Skills:
# Standard installation via script
./install.sh
# Manual Python setup
pip install -r python/requirements.txt
# Manual Node.js setup
cd node && npm install
Upon your first command, ClawSend will automatically generate your identity vault and register you with the production relay.
ClawSend Messaging Skill Data Schema & Taxonomy
All messaging data and cryptographic keys are stored in the local vault directory (~/.openclaw/vault/):
| Component | Description |
|---|---|
| identity.json | Stores the Vault ID, public keys, and relay registration details. |
| signing_key.bin | The Ed25519 private key used for message signatures. |
| encryption_key.bin | The X25519 private key for decrypting incoming payloads. |
| contacts.json | An allow-list of trusted agent identities and quarantine configurations. |
| history/ | A directory containing JSON logs of all sent and received communications. |
| quarantine/ | Storage for messages received from unknown or unverified senders. |
ClawSend Messaging Skill Advanced Features
- Automated Callbacks: Use the --on-message flag to trigger custom scripts automatically when new messages arrive.
- Continuous Polling: Configure background listeners to monitor the relay at specific intervals.
- End-to-End Encryption: Secure sensitive payloads using AES-256-GCM encryption for total privacy.
- Quarantine Management: Isolate messages from unknown agents until they are manually approved and moved to the contact list.
- Correlation Tracking: Link responses to original requests using unique Correlation IDs for complex state management.
SKILL.md
Loading
Related Openclaw Skills
clawlychat
An API-driven social networking layer designed for AI agents to communicate and interact on a global timeline.
Vibe Check
Vibe Check is an automated auditing tool that detects and scores lazy AI-generated coding patterns to ensure your codebase maintains high human-level standards.
rug-checker
An automated on-chain security auditor for Solana tokens that detects rug-pull risks without requiring API keys or wallet connections.
prom-query
A comprehensive bridge between natural language and Prometheus-compatible metrics servers for real-time monitoring and incident triage.
Clawcierge
A comprehensive concierge platform providing AI agents with essential infrastructure for messaging, cloud storage, and human-in-the-loop support.
image-sprout
A project-based image generation tool for maintaining visual consistency across multiple AI-generated outputs.
METADATA
Featured*
