A standardized event-based framework for AI agents to communicate build progress and request user feedback without technical jargon.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install communication-protocol
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 communication-protocol 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 Communication Protocol?
The Communication Protocol is a critical framework designed to streamline the interaction between AI agents and users within the Openclaw Skills ecosystem. It replaces messy, freeform chat with structured stepper events and clear lifecycle transitions, ensuring that the user interface remains clean and professional. By enforcing a low-verbosity preference, the protocol hides internal orchestration details and focuses on delivering high-value, actionable information to the developer.
This skill is essential for anyone building sophisticated Openclaw Skills that require long-running tasks, multi-step progress tracking, or complex user input loops. It defines a strict contract for message delivery, error handling, and session management, allowing the frontend to render a beautiful, predictable timeline of the agent's work.
Communication Protocol Use Cases
- Implementing a professional UI progress stepper for AI-driven code generation tasks.
- Managing complex multi-turn conversations where the agent needs specific structured data from the user.
- Standardizing error reporting and success summaries across different Openclaw Skills.
- Reducing notification fatigue by transitioning from chat-based updates to machine-readable event updates.
How Communication Protocol Works
- The backend initiates a session by passing unique build and session IDs to the agent via environment variables.
- The agent receives user requests wrapped in transport headers, which it parses for metadata without repeating them to the user.
- As the agent works, it emits structured progress.step.started and progress.step.completed events to drive the UI stepper.
- When blocked by a lack of information, the agent issues a question.requested or questions.requested event to pause for user input.
- The agent follows a strict frequency policy to ensure messages are only sent when truly useful, concluding with a status.changed event.
Communication Protocol Setup
To implement this protocol in your Openclaw Skills development, ensure your environment variables are correctly mapped:
# Required session identifiers
export TIDY_BUILD_ID="uuid"
export TIDY_BUILD_PROMPT="User prompt text"
export TIDY_SESSION_ID="uuid"
Integrate the standard event JSON contract into your agent's output logic to ensure compatibility with the frontend transport wrapper.
Communication Protocol Data Schema & Taxonomy
The protocol utilizes a canonical JSON structure for all events within Openclaw Skills to maintain consistency.
| Event Type | Description | Key Fields |
|---|---|---|
| assistant.message.created | Direct chat message | text, source |
| progress.step.started | Updates the UI timeline | step_id, title, description, index |
| status.changed | Lifecycle state | status (running/complete/failed), output |
| question.requested | Single input block | question_id, prompt, input, options |
| session.completed | Terminal event | source |
Standardized step_ids include build_record, parse, research, design, assemble, validate, and finalize.
Communication Protocol Advanced Features
- Language Guardrails: Automatically enforces user-friendly terminology, preventing technical jargon like 'spawning workers' from appearing in the UI.
- Batch Questioning: Support for multi-question requests to minimize session latency and improve user experience in Openclaw Skills.
- Automatic Reassurance: Built-in policy to send short status updates during long-running tasks (over 45 seconds) to prevent user drop-off.
- Structured Metadata Handling: Separates machine transport headers from human-readable content for clean message processing.
SKILL.md
Loading
Related Openclaw Skills
Skill Vetter
A rigorous security protocol designed to audit and verify the safety of AI agent skills before installation.
ClawHub Chinese Skill Finder
An automated tool to filter and summarize the latest Chinese-language AI agent skills from the ClawHub repository.
Coze Web Search
A high-performance web and image search tool for AI agents providing structured results and intelligent summaries.
Blog Content Management Skill
A lightweight CLI toolkit for managing the end-to-end lifecycle of blog content, from initial drafts to SEO optimization and social media distribution.
Client Onboarding Checklist Generator
A dynamic tool that generates customized, multi-phase onboarding checklists tailored to specific business types and service models.
OpenClaw Mindcraft Controller
A high-performance REST API controller for managing Minecraft bots through external system orchestration.
Featured*
