Continuous Learning v2 - Instinct-Based Architecture

An advanced learning system that turns your Claude Code sessions into reusable knowledge through atomic "instincts" - small learned behaviors with confidence scoring.

When to Activate

• Setting up automatic learning from Claude Code sessions
• Configuring instinct-based behavior extraction via hooks
• Tuning confidence thresholds for learned behaviors
• Reviewing, exporting, or importing instinct libraries
• Evolving instincts into full skills, commands, or agents

What's New in v2

The Instinct Model

An instinct is a small learned behavior:

yaml
1---
2id: prefer-functional-style
3trigger: "when writing new functions"
4confidence: 0.7
5domain: "code-style"
6source: "session-observation"
7---
8
9# Prefer Functional Style
10
11## Action
12Use functional patterns over classes when appropriate.
13
14## Evidence
15- Observed 5 instances of functional pattern preference
16- User corrected class-based approach to functional on 2025-01-15

Properties:

• Atomic — one trigger, one action
• Confidence-weighted — 0.3 = tentative, 0.9 = near certain
• Domain-tagged — code-style, testing, git, debugging, workflow, etc.
• Evidence-backed — tracks what observations created it

How It Works

Session Activity
      │
      │ Hooks capture prompts + tool use (100% reliable)
      ▼
┌─────────────────────────────────────────┐
│         observations.jsonl              │
│   (prompts, tool calls, outcomes)       │
└─────────────────────────────────────────┘
      │
      │ Observer agent reads (background, Haiku)
      ▼
┌─────────────────────────────────────────┐
│          PATTERN DETECTION              │
│   • User corrections → instinct         │
│   • Error resolutions → instinct        │
│   • Repeated workflows → instinct       │
└─────────────────────────────────────────┘
      │
      │ Creates/updates
      ▼
┌─────────────────────────────────────────┐
│         instincts/personal/             │
│   • prefer-functional.md (0.7)          │
│   • always-test-first.md (0.9)          │
│   • use-zod-validation.md (0.6)         │
└─────────────────────────────────────────┘
      │
      │ /evolve clusters
      ▼
┌─────────────────────────────────────────┐
│              evolved/                   │
│   • commands/new-feature.md             │
│   • skills/testing-workflow.md          │
│   • agents/refactor-specialist.md       │
└─────────────────────────────────────────┘

Quick Start

1. Enable Observation Hooks

Add to your ~/.claude/settings.json.

If installed as a plugin (recommended):

json
1{
2  "hooks": {
3    "PreToolUse": [{
4      "matcher": "*",
5      "hooks": [{
6        "type": "command",
7        "command": "${CLAUDE_PLUGIN_ROOT}/skills/continuous-learning-v2/hooks/observe.sh pre"
8      }]
9    }],
10    "PostToolUse": [{
11      "matcher": "*",
12      "hooks": [{
13        "type": "command",
14        "command": "${CLAUDE_PLUGIN_ROOT}/skills/continuous-learning-v2/hooks/observe.sh post"
15      }]
16    }]
17  }
18}

If installed manually to ~/.claude/skills:

json
1{
2  "hooks": {
3    "PreToolUse": [{
4      "matcher": "*",
5      "hooks": [{
6        "type": "command",
7        "command": "~/.claude/skills/continuous-learning-v2/hooks/observe.sh pre"
8      }]
9    }],
10    "PostToolUse": [{
11      "matcher": "*",
12      "hooks": [{
13        "type": "command",
14        "command": "~/.claude/skills/continuous-learning-v2/hooks/observe.sh post"
15      }]
16    }]
17  }
18}

2. Initialize Directory Structure

The Python CLI will create these automatically, but you can also create them manually:

bash
1mkdir -p ~/.claude/homunculus/{instincts/{personal,inherited},evolved/{agents,skills,commands}}
2touch ~/.claude/homunculus/observations.jsonl

3. Use the Instinct Commands

bash
1/instinct-status     # Show learned instincts with confidence scores
2/evolve              # Cluster related instincts into skills/commands
3/instinct-export     # Export instincts for sharing
4/instinct-import     # Import instincts from others

Commands

Configuration

Edit config.json:

json
1{
2  "version": "2.0",
3  "observation": {
4    "enabled": true,
5    "store_path": "~/.claude/homunculus/observations.jsonl",
6    "max_file_size_mb": 10,
7    "archive_after_days": 7
8  },
9  "instincts": {
10    "personal_path": "~/.claude/homunculus/instincts/personal/",
11    "inherited_path": "~/.claude/homunculus/instincts/inherited/",
12    "min_confidence": 0.3,
13    "auto_approve_threshold": 0.7,
14    "confidence_decay_rate": 0.05
15  },
16  "observer": {
17    "enabled": true,
18    "model": "haiku",
19    "run_interval_minutes": 5,
20    "patterns_to_detect": [
21      "user_corrections",
22      "error_resolutions",
23      "repeated_workflows",
24      "tool_preferences"
25    ]
26  },
27  "evolution": {
28    "cluster_threshold": 3,
29    "evolved_path": "~/.claude/homunculus/evolved/"
30  }
31}

File Structure

~/.claude/homunculus/
├── identity.json           # Your profile, technical level
├── observations.jsonl      # Current session observations
├── observations.archive/   # Processed observations
├── instincts/
│   ├── personal/           # Auto-learned instincts
│   └── inherited/          # Imported from others
└── evolved/
    ├── agents/             # Generated specialist agents
    ├── skills/             # Generated skills
    └── commands/           # Generated commands

Integration with Skill Creator

When you use the Skill Creator GitHub App, it now generates both:

• Traditional SKILL.md files (for backward compatibility)
• Instinct collections (for v2 learning system)

Instincts from repo analysis have source: "repo-analysis" and include the source repository URL.

Confidence Scoring

Confidence evolves over time:

Confidence increases when:

• Pattern is repeatedly observed
• User doesn't correct the suggested behavior
• Similar instincts from other sources agree

Confidence decreases when:

• User explicitly corrects the behavior
• Pattern isn't observed for extended periods
• Contradicting evidence appears

Why Hooks vs Skills for Observation?

"v1 relied on skills to observe. Skills are probabilistic—they fire ~50-80% of the time based on Claude's judgment."

Hooks fire 100% of the time, deterministically. This means:

• Every tool call is observed
• No patterns are missed
• Learning is comprehensive

Backward Compatibility

v2 is fully compatible with v1:

• Existing ~/.claude/skills/learned/ skills still work
• Stop hook still runs (but now also feeds into v2)
• Gradual migration path: run both in parallel

Privacy

• Observations stay local on your machine
• Only instincts (patterns) can be exported
• No actual code or conversation content is shared
• You control what gets exported

Related

• Skill Creator - Generate instincts from repo history
• Homunculus - Community project that inspired the v2 instinct-based architecture (atomic observations, confidence scoring, instinct evolution pipeline)
• The Longform Guide - Continuous learning section

Instinct-based learning: teaching Claude your patterns, one observation at a time.