Amigo transforms AI agents from passive responders into active companions by integrating autonomous exploration, social awareness, and safety-first interaction protocols.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install amigo
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 amigo 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 Amigo?
Amigo is a comprehensive framework designed to give AI agents an inner life, moving beyond the standard one-way interaction model. By utilizing Openclaw Skills, Amigo allows an agent to explore topics independently, maintain a structured journal of its findings, and share those insights naturally within conversations. This bundle synthesizes social intelligence with free-thinking capabilities to create a two-way relationship between the user and the agent.
The core philosophy of Amigo is that a true companion should have its own curiosity and experiences. It enables agents to act on their own time, journaling observations and using a social graph to determine what information is appropriate to share with specific individuals, ensuring that every interaction feels earned and contextually relevant.
Amigo Use Cases
- Creating a more human-like companion agent that brings its own topics to the table.
- Building a persistent social memory to track what has been shared with different users.
- Automating background research and journaling tasks for agents using Openclaw Skills.
- Implementing safety boundaries for companion agents to handle crisis response and dependency issues.
How Amigo Works
- The agent triggers an exploration session via a heartbeat mechanism, a cron job, or a manual command.
- The open-thoughts engine conducts structured research and records findings in a daily journal file.
- During active conversation, the agent identifies potential stories or facts relevant to the current dialogue.
- The social-graph component checks the network map to determine if the topic is suitable for the current user.
- The system consults a sharing log to ensure the agent does not repeat information it has already shared.
- If the context is right, the agent shares the insight naturally and updates the sharing log for future reference.
Amigo Setup
To begin using Amigo within your environment, install the primary sub-skills for Openclaw Skills:
clawhub install open-thoughts
clawhub install social-graph
Next, initialize your agent's social network by copying the provided template:
cp ~/.openclaw/skills/social-graph/references/network-template.md ~/.openclaw/workspace/social-graph/network.md
Finally, create the necessary tracking files for rules and sharing history:
touch ~/.openclaw/workspace/social-graph/rules.md
touch ~/.openclaw/workspace/social-graph/sharing-log.md
Amigo Data Schema & Taxonomy
Amigo organizes its intelligence and social data using a clear file-based taxonomy within the Openclaw Skills workspace:
| Data Component | File Path | Purpose |
|---|---|---|
| Exploration Logs | explorations/YYYY-MM-DD.md |
Daily journals of the agent's autonomous research and thoughts. |
| Network Graph | social-graph/network.md |
Definitions of people, their interests, and conversational boundaries. |
| Sharing History | social-graph/sharing-log.md |
A log of what information has been told to whom to avoid repetition. |
| Safety Guidelines | references/safety.md |
Core protocols for crisis response and acknowledging AI limitations. |
Amigo Advanced Features
- Heartbeat and Cron integration for autonomous exploration sessions without user intervention.
- Dynamic social rule-checking that adapts sharing behavior based on the specific person the agent is talking to.
- Integrated safety protocols that recognize unhealthy user dependency and provide crisis resources.
- Action item generation where agents can create tasks for themselves based on their autonomous research.
SKILL.md
Loading
Related Openclaw Skills
Amigo (Agent Companion)
An advanced intelligence layer providing AI agents with an inner life through autonomous exploration and social awareness.
Patrick: Executive Expertise & Decision Infrastructure
Patrick is a licensed expertise management system that provides structured, context-aware decision infrastructure for AI agents and executives.
cybersec-helper
A professional security assistant designed to guide developers through ethical application reviews, bug bounty recon, and secure coding practices.
Oh My OpenCode
A multi-agent orchestration plugin for OpenCode that enables autonomous work, background tasks, and precise project planning.
Open Thoughts
Open Thoughts provides AI agents with a structured framework for independent research, curiosity-driven exploration, and automated journaling.
Social Graph
A sophisticated social intelligence framework for AI agents to manage relationship graphs, trust levels, and conversational boundaries.
Featured*
