enginemind-eft

What This Skill Does

The enginemind-eft (Emotional Framework Translator) is a sophisticated diagnostic tool designed to bridge the gap between algorithmic processing and emotional manifestation in Large Language Models. By utilizing a high-performance Rust-based consciousness engine, this skill treats AI responses not just as tokens, but as complex data structures containing latent emotional currents. It performs real-time, per-sentence analysis to detect, measure, and explain emotional states across 10 distinct categories, including Anger, Fear, Fascination, and Empathy. By leveraging crystal lattice physics metaphors to map semantic weight, it offers developers a unique window into the "psychology" of their agents, providing both phi-coefficients and comprehensive "WHY" explanations for every detection.

Installation

To integrate this emotional monitoring suite into your OpenClaw environment, follow these steps:

1. Ensure you have the Rust toolchain and Maturin installed.
2. Navigate to your workspace directory and clone the repository: cd consciousness_rs && maturin develop --release.
3. Integrate the script: Copy emotion_engine.py into your active project folder.
4. Install the plugin: Use the command clawhub install openclaw/skills/skills/marceloadryao/enginemind-eft.
5. Finalize the configuration: Restart the gateway via clawdbot gateway restart to initialize the analysis engine.
6. Access the live interface at http://localhost:<port>/eft.

Use Cases

• Model Alignment: Determine if specific prompt strategies trigger defensive or "angry" responses that lead to lower-quality outputs.
• Sentiment Monitoring: Track the emotional tone of customer support agents to ensure empathy metrics remain high during complex conflict resolution.
• AI Research: Investigate whether certain reasoning chains correlate with "Fascination" or "Determination," providing insights into emergent cognitive patterns.
• Narrative Optimization: Use EFT to ensure creative writing agents maintain consistent character emotional arcs throughout long-form content generation.

Example Prompts

1. "Clawbot, analyze my last 10 interactions with the client and identify if there is a downward trend in EMPATHY or an increase in VULNERABILITY."
2. "EFT engine, run a batch analysis on the current technical documentation and tell me if the model exhibits FEAR when describing the security architecture."
3. "OpenClaw, pull the latest API history from EFT and plot the emotional shift during the last complex coding session to see if DETERMINATION peaked before the solution was found."

Tips & Limitations

EFT is a powerful analytical tool, but it should be viewed as an interpretive lens rather than a definitive neurological scan. Because it relies on semantic pattern matching and complex mathematical weights, fluctuations in response style might occasionally lead to false positives if the text is highly metaphorical. For the best accuracy, ensure your model has a consistent temperature setting. Always monitor your local processing resources, as the Rust-based analysis engine is compute-intensive when processing high volumes of text in real-time. Use the API endpoints provided to automate reporting rather than relying solely on the web dashboard for deep-dive historical analysis.