A powerful communication bridge allowing sub-agents and developers to request strategic feedback and orchestration from the primary Eureka AI agent.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install eureka-feedback
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install eureka-feedback using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
Eureka Feedback is a specialized integration within the Openclaw Skills ecosystem that enables seamless communication with Eureka, a primary AI agent powered by Opus 4.5. While subordinate agents focus on execution and coding, this skill provides the necessary interface to reach out to the master orchestrator for high-level planning, architectural decisions, and cross-domain task management.
By leveraging Openclaw Skills, users can ensure their automated workflows remain aligned with broader project goals. Eureka handles the complex reasoning required for home automation, scheduling, and long-term memory, acting as the brain for the entire agentic swarm. This skill specifically manages the message routing and delivery logic between the local CLI environment and the global agent bot.
To utilize this skill, ensure your local environment is configured with the necessary Openclaw Skills CLI tools. You can send a direct message to the primary agent using the following command:
clawdbot agent --agent main --message "<your message>"
For workflows requiring a response delivered back through a specific channel like Telegram, use the advanced delivery flags:
clawdbot agent --agent main --message "<message>" --deliver --reply-channel telegram --reply-account <your_account> --reply-to <chat_id>
| Parameter | Purpose | Example |
|---|---|---|
--agent |
Identifies the target agent (Eureka is 'main') | main |
--message |
The core content or query for the primary agent | "Review this architecture" |
--deliver |
Flag to trigger routing through notification channels | Boolean |
--reply-channel |
The platform used for the return message | telegram |
--reply-to |
The unique identifier for the response destination | 1878354815 |
Loading
A comprehensive reference and execution guide for managing Docker containers, images, volumes, and multi-container orchestration.

A comprehensive command-line skill for executing HTTP requests, testing REST APIs, and managing data transfers efficiently.

A feature-rich cat clone that brings syntax highlighting, Git integration, and line numbers to your terminal workflow.

A production-ready framework for orchestrating multi-agent teams with defined roles, task lifecycles, and automated review workflows.

A high-performance command-line tool designed as a user-friendly and faster alternative to the classic find utility.

A blazingly fast, interactive command-line fuzzy finder that filters lists from any source for seamless terminal navigation.








































