A specialized skill for programmatically interacting with CoPaw instances via HTTP API to manage agents, sessions, and real-time SSE communication.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install copaw-api-chat
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 copaw-api-chat 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 CoPaw API Chat?
The CoPaw API Chat skill is designed for developers who need to move beyond the web UI and interact with CoPaw instances programmatically. It provides a structured framework for navigating the complex HTTP API surface, ensuring that agents are correctly scoped and sessions are managed effectively through the Openclaw Skills ecosystem.
By leveraging this skill, developers can automate agent interactions, build custom wrappers, and integrate real-time AI responses into external applications. It emphasizes the importance of understanding Server-Sent Events (SSE) and session-centric communication, moving away from stateless request/response patterns toward robust, context-aware AI integrations.
CoPaw API Chat Use Cases
- Programmatically interacting with a local CoPaw instance for automated testing or integration.
- Building custom frontend wrappers or dashboards on top of the CoPaw backend.
- Automating the creation of chats and sessions for multi-agent workflows.
- Investigating CoPaw authentication layers and agent scoping before deploying production code.
- Monitoring agent outputs and system states via real-time SSE stream consumption.
How CoPaw API Chat Works
- Identity Confirmation: The skill first identifies the target agent ID, defaulting to the system default unless specific overrides are detected.
- Session Initialization: It executes a POST request to the chat endpoint to establish a unique session and user context.
- Context Injection: Necessary headers (like X-Agent-Id) and session tokens are captured to maintain stateful communication.
- Message Dispatch: A payload is sent to the agent-specific console endpoint, triggering the AI's processing logic.
- Stream Processing: The skill interprets the resulting Server-Sent Events (SSE) to deliver real-time, incremental updates to the user or calling system.
CoPaw API Chat Setup
To start using this skill within the Openclaw Skills environment, ensure your CoPaw instance is reachable and configure your local environment variables for authentication:
# Example: Setting up Basic Auth if required by your proxy
export COPAW_API_KEY="your_api_key_here"
export COPAW_BASE_URL="http://localhost:8080"
Refer to the internal documentation at references/overview-auth-scoping.md to confirm if your deployment requires Nginx Basic Auth or internal bearer tokens.
CoPaw API Chat Data Schema & Taxonomy
The skill follows a strict hierarchical data taxonomy to ensure session integrity across the Openclaw Skills framework:
| Object | Description | Key Fields |
|---|---|---|
| Chat Session | The root container for a conversation | session_id, user_id, channel |
| Agent Scope | Defines the behavior and tools available | agentId, X-Agent-Id header |
| SSE Stream | Real-time event data format | event, data, id |
| Auth Layer | Security configuration | Basic Auth or Internal Auth toggles |
CoPaw API Chat Advanced Features
- Multi-agent scoping using the agent-scoped route system (
/api/agents/{agentId}/...). - Advanced workspace management for handling file surfaces and admin-level APIs.
- Integration with MCP (Model Context Protocol) and cron boundaries for scheduled agent tasks.
- Specialized handling for SSE mechanics to manage long-running generative responses.
- High-level wrappers for complex operations like tool management and model swapping.
SKILL.md
Loading
Related Openclaw Skills
XiaoHongShu Content Creation Skill
An advanced AI-driven content engine designed to craft, optimize, and stylize posts specifically for the XiaoHongShu platform.
Qimen Dunjia Chart Analyzer
A deterministic Qimen Dunjia calculation and interpretation engine for strategic timing and spatial forecasting.
EvalLayer Evaluator
An AI-powered evaluation gate that extracts factual claims and scores deliverable quality for autonomous agent workflows.
Meta Workflow Discoverer
An intelligent pattern-recognition engine that observes user behavior to automatically discover and generate executable automation workflows.
zkSync Reference Tool
A comprehensive CLI-based reference suite for developers and auditors working within the zkSync blockchain ecosystem.
Task Executor
Task Executor is an advanced autonomous agent skill that transforms high-level user requirements into completed projects through automated task decomposition and multi-agent coordination.
Featured*
