claude-assess-decide-do-mega-prompt: Skill for Claude Code

ADD Flow Check

Model-invocable skill for automatic flow status updates.

---

description: Automatically check and update ADD flow status at conversation boundaries model-invocable: true user-invocable: false

Purpose

This skill enables Claude to automatically monitor ADD realm patterns and update the .add-status file at natural conversation boundaries. It runs silently in the background, maintaining flow observability without user intervention.

When to Invoke

Claude should invoke this skill automatically at:

• Realm transitions - When linguistic patterns indicate shift between Assess/Decide/Do
• Significant pattern shifts - Analysis paralysis emerging, decision clarity gathering, etc.
• Completion events - Tasks finished, decisions made, livelines created
• Every 5-7 exchanges if no boundary occurs (ambient awareness)

Realm Detection Patterns

Assess Realm Indicators

• "I'm thinking about...", "What are my options for...", "Help me understand..."
• "What if I...", "I'm not sure yet, but...", exploratory questions
• Future/conditional language: "might", "could", "would if", "considering"
• Low commitment: "maybe", "thinking about", "not sure"

Decide Realm Indicators

• "Should I...", "I need to choose between...", "When should I..."
• "What's the priority...", "I want to commit to..."
• Present/committed language: "I want to", "I'm choosing", "This matters"
• Comparison/weighing: "Which is better?", "What matters here?"

Do Realm Indicators

• "How do I actually...", "I need to complete...", "Walk me through the steps..."
• "I'm working on...", active execution language
• Immediate language: "I'm doing", "working on", "completing", "finished"
• Action requests and completion acknowledgments

Status File Format

Update .add-status with pipe-delimited format:

REALM|EMOJI|PATTERN_DESCRIPTION|EXCHANGES|TRANSITIONS

Emoji Reference

• Assess: 🔴+ (red, adding data)
• Decide: 🟠? (orange, choosing)
• Do: 🟢- (green, executing/removing)