Kandy is a powerful real-time communications platform integration that enables AI agents to manage voice, video, and messaging workflows seamlessly.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install kandy
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 kandy 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 Kandy?
The Kandy skill for Openclaw Skills provides a robust bridge to the Kandy Communications Platform as a Service (CPaaS). By leveraging this skill, developers can empower their AI agents to interact with real-time communication features such as call history, voicemail, and conversation management without manually handling complex API authentication.
This integration uses Membrane as the underlying engine to automate credential management and token refreshes. This ensures that your Openclaw Skills remains focused on the logic of communication and customer engagement rather than the plumbing of OAuth or secret storage, providing a secure and scalable way to embed telecommunications into AI-driven workflows.
Kandy Use Cases
- Automating customer engagement by responding to or initiating Kandy conversations.
- Syncing call history and voicemail data into external CRM or database systems.
- Building AI-driven voice and video collaboration tools for internal team communication.
- Programmatic retrieval of communication records for data analytics and reporting.
- Scaling CPaaS workflows within headless environments using streamlined CLI commands.
How Kandy Works
- Install the Membrane CLI to provide the underlying infrastructure for Openclaw Skills connectivity.
- Authenticate the agent using the login command to establish a secure tenant session.
- Create or ensure a connection to the Kandy domain, which triggers the automated authentication flow.
- Poll the connection status until it reaches a READY state, indicating that Membrane has finalized the auth handshake.
- Search for specific Kandy actions using natural language intents to find pre-built tools for messaging or calls.
- Execute actions or proxy raw requests to the Kandy API while Membrane handles all header injection and credential refreshes.
Kandy Setup
1. Install the Membrane CLI
Install the CLI globally to enable Openclaw Skills to communicate with Kandy:
npm install -g @membranehq/cli@latest
2. Authenticate the Agent
Login and specify openclaw as the agent type to optimize the tooling:
membrane login --tenant --clientName=openclaw
3. Establish a Kandy Connection
Connect to the Kandy platform to initialize the managed auth lifecycle:
membrane connection ensure "https://www.kandy.io/" --json
Kandy Data Schema & Taxonomy
The Kandy skill organizes data through a structured connection and action model. All interactions return standardized JSON objects for high compatibility with Openclaw Skills.
| Data Type | Description | Key Fields |
|---|---|---|
| Connection | The auth state and metadata for the Kandy link. | id, state, clientAction |
| Action | Pre-defined functions like 'List Calls' or 'Send Message'. | id, inputSchema, outputSchema |
| Request | Raw proxy data returned from direct Kandy API endpoints. | output, status, headers |
| Metadata | Information about the skill version and authorship. | version, author, homepage |
Kandy Advanced Features
- Natural Language Action Discovery: Find the exact communication tool needed by describing the intent.
- Transparent Credential Refresh: Openclaw Skills never lose access thanks to server-side token management.
- Headless Environment Support: Complete authentication in restricted environments via authorization codes.
- Direct API Proxying: Access any Kandy endpoint not covered by pre-built actions with automatic header injection.
- Multi-Agent Optimization: Adjust tooling behavior by specifying agent types during the login process.
SKILL.md
Loading
Related Openclaw Skills
Retail Sales Performance Analysis
A professional business intelligence skill for retail environments that provides automated period-over-period sales analysis and performance diagnostics.
Frontend Code Review
An automated auditing tool that reviews frontend code for architecture, type safety, security, and accessibility, generating detailed Markdown reports.
Complex Task Executor
A structured methodology for AI agents to plan, execute, and review complex operations while automatically persisting lessons learned into long-term memory.
Weekly Report Skill
An automated reporting agent that aggregates data and publishes structured weekly reports directly to Enterprise WeChat (WeCom) documents.
Obsidian Vault Context
A skill that transforms an Obsidian vault into a shared, persistent workspace where AI agents and users collaborate via markdown files rather than just chat.
SellerSprite Product Research
A powerful command-line interface for SellerSprite that enables automated Amazon market research, keyword analysis, and competitor tracking.
Featured*
