continuous-learning-v2

Transform Gemini CLI sessions into atomic instincts with confidence scoring.

Updated Mar 19, 2026
One-click install
npx skills add https://github.com/espensev/ai-skills --skill continuous-learning-v2-espensev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: continuous-learning-v2
Source: https://github.com/espensev/ai-skills/tree/main/gemini-skills/skills/continuous-learning-v2
Command: npx skills add https://github.com/espensev/ai-skills --skill continuous-learning-v2-espensev

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Instinct-based learning turns Gemini CLI sessions into reusable knowledge through atomic "instincts" - small learned behaviors with confidence scoring, and supports project-scoped instincts to prevent cross-project contamination.

Core Features & Use Cases

  • Atomic instincts with confidence scoring that can evolve into skills, commands, or agents.
  • Hooks-based observation to capture prompts and tool usage for durable pattern discovery.
  • Project-scoped vs global instincts for isolation and targeted learning within a repository.

Quick Start

Enable observation hooks and initialize the instincts directory to begin capturing patterns from your Gemini sessions.

Frequently Asked Questions about continuous-learning-v2

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I capture reusable knowledge from Gemini CLI sessions?

Capture reusable knowledge from Gemini CLI sessions by enabling hooks-based observation to monitor prompts and tool usage, creating atomic instincts with confidence scoring for durable pattern discovery.

What are project-scoped instincts and how do they prevent cross-project contamination?

Project-scoped instincts are isolated learned behaviors tied to a specific repository, preventing cross-project contamination by keeping global and project-level pattern evolution separate.

How do atomic instincts evolve into skills, commands, or agents?

Atomic instincts evolve into skills, commands, or agents through an evolution pipeline with guardrails and promotion mechanisms that elevate high-confidence observations into actionable behaviors.

Can I apply globally learned instincts across multiple projects?

You can apply globally learned instincts across multiple projects by using the global context scope, allowing the system to reuse durable behaviors outside of isolated project-scoped environments.

How do I initialize the instincts directory to start capturing patterns?

Initialize the instincts directory by enabling observation hooks, which establishes the required project-scoped or global file structure to begin persisting observations and capturing session patterns.

What is the difference between project-scoped and global instincts for workflow learning?

Project-scoped instincts isolate workflow learning within a single repository to prevent cross-contamination, while global instincts allow reusable behaviors to be applied across multiple projects simultaneously.