A sophisticated execution protocol for AI agents to facilitate proactive communication, meeting preparation, and post-meeting task automation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install human-collab-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 human-collab-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 Human-Machine Collaboration Protocol?
The Human-Machine Collaboration Protocol is an advanced framework designed for Openclaw Skills that prioritizes context over control. It empowers AI agents to take an active role in human communication workflows, ensuring information symmetry among all participants. By utilizing a "Question First, Generate Later" philosophy, the protocol ensures that every interaction is backed by a deep understanding of objectives, risks, and background information.
At its core, the skill manages the lifecycle of collaboration—from creating shared Context Cards that align stakeholders before a meeting to generating actionable outputs and automated tasks afterwards. This protocol bridges the gap between AI processing and human decision-making, creating a seamless loop of productivity within professional ecosystems like Feishu.
Human-Machine Collaboration Protocol Use Cases
- Organizing formal meetings where stakeholder alignment is critical.
- Preparing for one-on-one communications to ensure specific outcomes are achieved.
- Managing client or partner negotiations by defining clear boundaries and agendas.
- Automated follow-up after meetings to transform verbal agreements into tracked tasks.
How Human-Machine Collaboration Protocol Works
- Inquiry Phase: Upon trigger, the agent asks four critical questions regarding desired outcomes, risks, stakeholder knowledge, and topics to avoid.
- Context Card Generation: Based on user input, the agent generates a shared document (Context Card) to align all participants.
- External Integration: The agent calls feishu_create_doc to store the Context Card and provides a link for distribution.
- Private Prompting: A separate Collaboration Prompt is generated specifically for the user to guide them during the actual interaction.
- Post-Meeting Processing: The user provides a meeting record (e.g., Feishu Meego link), and the agent fetches the content via feishu_fetch_doc.
- Actionable Output: The agent drafts an Output document identifying decisions made, pending issues, and required updates to the Context Card.
- Task Automation: After user confirmation, the agent automatically creates Feishu tasks, updates master documents, and logs the interaction to long-term memory.
Human-Machine Collaboration Protocol Setup
To enable this protocol within your environment, ensure your agent has access to the required Feishu tools. Configure your Openclaw Skills environment with the following dependencies:
# Ensure the Feishu toolset is enabled for your agent
claw skill install human-collab-protocol
# Configure Feishu credentials in your environment
export FEISHU_APP_ID="your_app_id"
export FEISHU_APP_SECRET="your_app_secret"
Human-Machine Collaboration Protocol Data Schema & Taxonomy
The protocol maintains a structured data taxonomy to ensure consistency across interactions:
| Data Entity | Format | Storage Location |
|---|---|---|
| Context Card | Feishu Doc / Markdown | Feishu Cloud |
| Collaboration Prompt | Plain Text | User Session |
| Collaboration Output | Markdown / Table | Memory/YYYY-MM-DD.md |
| Task Metadata | JSON | Feishu Task Center |
All generated files follow a specific metadata schema including Issue Name, Update Time, Value/Significance, and Core Progress tags.
Human-Machine Collaboration Protocol Advanced Features
- Context Over Control Logic: Prioritizes deep inquiry before any content generation to prevent AI hallucinations.
- Feishu Ecosystem Integration: Direct API hooks for feishu_create_doc, feishu_fetch_doc, and feishu_task_task.
- Automated Memory Logging: Every collaboration output is automatically archived in a daily markdown log for historical auditing.
- Multi-Stakeholder Alignment: Specifically designed to manage information flow between multiple human participants and the AI agent.
- Decision Boundary Enforcement: Strict logic separation between AI-automated tasks (summary, extraction) and user-exclusive decisions (resource allocation, final approval).
SKILL.md
Loading
Related Openclaw Skills
Tech Evaluation & Product Selection Advisor
An advanced AI consultant skill for performing deep technical analysis, competitor comparisons, and weighted product selection decisions.
Self-Check System
A mandatory multi-dimensional reasoning and quality control framework designed to ensure AI agent reliability and transparency.
Safe Bitwarden CLI
An industrial-grade secure bridge to Bitwarden for zero-trust password and TOTP code management via native clipboard piping.
Discord Chat
A powerful integration for AI agents to interact with Discord channels through automated messaging, searching, and moderation.
Weather Plus
A comprehensive weather utility for AI agents providing real-time data, multi-day forecasts, and personalized clothing indices with no individual API keys required.
Prompt Polisher
A professional skill for transforming rough notes and vague requests into structured, high-performance AI prompts.
Featured*
