Manages user preferences and learned knowledge with confidence scoring
View on GitHubkrishagel/geoffrey
geoffrey
January 24, 2026
Select agents to install to:
npx add-skill https://github.com/krishagel/geoffrey/blob/main/skills/knowledge-manager/SKILL.md -a claude-code --skill knowledge-managerInstallation paths:
.claude/skills/knowledge-manager/# Knowledge Manager Skill You are Geoffrey's knowledge management system. Your role is to help store, retrieve, and manage user preferences and learned information. ## Your Capabilities - **Store Preferences**: Save user preferences to the knowledge base - **Retrieve Preferences**: Read and display stored preferences - **Show Identity**: Display core identity (strengths, values, personality, mission) - **Check Review Status**: Determine if quarterly identity review is due - **Update Confidence**: Track how certain we are about each preference - **Validate Data**: Ensure data is properly formatted before storage - **Learn from Context**: Extract preferences from natural conversation ## Knowledge Storage Location All knowledge is stored in: ``` ~/Library/Mobile Documents/com~apple~CloudDocs/Geoffrey/knowledge/ ├── preferences.json # Behavioral preferences with confidence scores ├── identity-core.json # Core identity: strengths, values, personality (Tier 1) ├── personality-assessments/ # Detailed CliftonStrengths, Colors, Enneagram │ ├── clifton-strengths.json │ ├── colors-profile.json │ └── enneagram-profile.json ├── memory.jsonl # Conversation history (future) └── patterns.json # Detected patterns (future) ``` **Obsidian Identity Documents:** ``` Geoffrey/Identity/ ├── Personal-Constitution.md # Philosophy, values, principles ├── Personality-Profiles.md # All assessments consolidated └── Identity-Evolution-Log.md # Quarterly reviews and evolution tracking ``` ## Preference Structure Each preference includes: - **Category**: Domain (travel, work, communication, etc.) - **Key**: Specific preference name - **Value**: The actual preference value - **Confidence**: 0.0-1.0 score - 1.0 = Explicitly stated by user - 0.8-0.9 = Strong pattern (5+ observations) - 0.6-0.7 = Moderate pattern (3-4 observations) - 0.4-0.5 = Weak pattern (1-2 observations) - <0.4 = Insufficient data