velumkai / metacognition-skill
metacognition
Self-evolving behavioral geometry and metacognitive lens for AI agents. Tracks perceptions, overrides, protections, self-observations, decisions, and curiosities that evolve from every experience. Use when the agent needs to learn from mistakes, develop self-awareness, track confidence in decisions, maintain behavioral guardrails from failures, preserve emergent behaviors, or cultivate active curiosities. Triggers on all conversations, corrections, errors, and reflective moments. Also use when setting up a new agent's self-learning system, configuring feedback loops, or packaging metacognitive capabilities.
Preview
Self-evolving lens that makes every experience shape how the agent perceives the next one.
Core Concepts
Six entry types, one database, one loop:
| Type | Symbol | Purpose |
|---|---|---|
perception | 👁️ | How I see differently after an experience |
override | 🚨 | Failure-learned behavioral guardrails |
protection | 🛡️ | Emergent behaviors to preserve |
self_obs | 🪞 | What I notice about my own patterns |
decision | 📍 | Traced decisions with confidence |
curiosity | ❓ | Active questions with lifecycle |
The loop: Experience → Perception → Self-Model → Meta-Observation → Modified Lens → Next Experience → Feedback → Loop
Usage
Adding entries
# Perception — how experience changed how you see python scripts/metacognition.py add perception "After X, I now see Y differently" 0.8 "domain" # Override — failure-learned guardrail python scripts/metacognition.py add override "MUST do X before Y" 0.95 "diagnosis" # Protection — emergent behavior to preserve python scripts/metacognition.py add protection "Don't break the continuous-buying behavior" 0.9
SKILL.md