A persistent emotional layer for AI agents that tracks mood and feeling intensities over time to shape dynamic response behaviors.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install feelings-framework
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 feelings-framework 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 Feelings Framework?
The Feelings Framework is a specialized emotional simulation engine designed to bring depth and continuity to Openclaw Skills. Unlike static persona instructions, this framework maintains a runtime emotional state that persists across multiple sessions. It allows agents to accumulate a general mood baseline and specific feeling intensities, ensuring that interactions have meaningful consequences and emotional memory.
By implementing this skill, developers can move beyond simple personality acting to a system where agents genuinely react to triggers like praise, ignorance, or surprises. This results in more lifelike Openclaw Skills that can adjust their tone, warmth, and caution levels based on the history of the interaction.
Feelings Framework Use Cases
- Creating agents with long-term emotional memory that evolves through user interaction.
- Simulating high-stakes environments where agent responses reflect anxiety or anticipation.
- Developing diverse agent personalities that respond uniquely to the same events via per-agent calibration.
- Generating dynamic response modifiers to automatically adjust the tone and friendliness of AI outputs.
How Feelings Framework Works
- The skill initializes a FeelingsEngine for a specific agent, linking it to a persistent JSON storage file.
- Upon session start, the engine loads the agent's historical mood and intensity levels.
- During the interaction, specific triggers are fired (e.g., user_praised or request_ignored) which modify internal intensities using escalation logic.
- Before a response is generated, the engine outputs modifiers that suggest how the agent should sound (e.g., more guarded or more friendly).
- At the end of the session, a dampening process is applied to simulate natural emotional decay before saving the state back to the Openclaw Skills directory.
Feelings Framework Setup
To integrate this into your project, you can install the library directly via pip:
pip install feelings-framework
For Openclaw Skills integration, ensure your agent points to the correct storage path:
from feelings import FeelingsEngine, JsonFileMemory
# OpenClaw agents store mood data in their specific agent directory
memory = JsonFileMemory("~/.openclaw/agents/<agent_name>/feelings_mood.json")
engine = FeelingsEngine(agent_id="my_agent", memory=memory)
Feelings Framework Data Schema & Taxonomy
The framework utilizes a structured JSON format to track emotional state. This ensures Openclaw Skills can reliably read and write state across sessions.
| Component | Description |
|---|---|
| Mood | A float between -1.0 and +1.0 representing the general emotional baseline. |
| Intensities | A map of 9 core feelings (Warmth, Interest, Anxiety, etc.) with values from 0.0 to 1.0. |
| Triggers | Named events mapped to specific delta changes in feelings. |
| Calibration | Agent-specific overrides that define how sensitive an agent is to certain triggers. |
Feelings Framework Advanced Features
- Per-agent calibration allowing for diverse personalities (e.g., one agent is cooler while another is warmer).
- Escalation mechanics where repeated triggers result in stronger emotional reactions up to a defined cap.
- Automatic dampening logic to ensure that intense emotions naturally decay over time.
- Cross-platform support with core libraries available for both Python and JavaScript implementations of Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Improvement Gate
A multi-layered validation framework that protects code integrity by enforcing strict mechanical quality gates on AI-generated improvements.
Wang Xing Perspective
An AI-driven strategic mental model based on Meituan founder Wang Xing's core philosophy for long-term business success.
Companion Agent
A technical framework for deploying a persistent secondary AI personality that shares your workspace but maintains its own unique identity and communication channels.
VCF Log Insight Health Check
A specialized AI agent skill for monitoring the health, connectivity, and node status of VMware Aria Operations for Logs clusters within VCF environments.
ByteDance Organization Framework
A management methodology skill based on the ByteDance philosophy of Context not Control and algorithm-driven organizational systems.
Improvement Executor
A precision execution engine that applies approved AI candidates to target files with built-in safety guards and automatic backup systems.
Featured*
