Token Budget

Token Budget Skill

Core principle: Every token counts. Load only what you need, when you need it.

Token Estimation

Quick Estimates

Rule of thumb: tokens ≈ lines × 4

File Size Categories

Budget Thresholds

Based on PROJECT_RULES.md context quality thresholds:

Budget Tracking Protocol

Before Each Task

1. Estimate current usage:

   • Count files in context
   • Estimate tokens per file
   • Calculate approximate %

2. Check budget status:

   Current: ~X,000 tokens (~Y%)
   Budget: [PEAK|GOOD|DEGRADING|POOR]
   

3. Adjust strategy:

   • PEAK: Proceed normally
   • GOOD: Prefer search-first
   • DEGRADING: Use outlines only
   • POOR: Trigger state dump

During Execution

Track cumulative context:

## Token Tracker

| Phase | Files Loaded | Est. Tokens | Cumulative |
|-------|--------------|-------------|------------|
| Start | 0 | 0 | 0 |
| Task 1 | 2 | ~400 | ~400 |
| Task 2 | 3 | ~600 | ~1000 |

Optimization Strategies

1. Progressive Loading

Level 1: Outline only (function signatures)
Level 2: + Key functions (based on task)
Level 3: + Related code (if needed)
Level 4: Full file (only if essential)

2. Just-In-Time Loading

• Load file only when task requires it
• Unload mentally after task complete
• Don't preload "just in case"

3. Search Before Load

Always use context-fetch skill first:

1. Search for relevant terms
2. Identify candidate files
3. Load only needed sections

4. Summarize & Compress

After understanding a file:

• Document key insights in STATE.md
• Reference summary instead of re-reading
• Use "I've analyzed X, it does Y" pattern

Budget Alerts

At 50% Budget

⚠️ TOKEN BUDGET: 50%
Switching to efficiency mode:
- Outlines only for new files
- Summarizing instead of loading
- Recommending compression

At 70% Budget

🛑 TOKEN BUDGET: 70%
Quality degradation likely. Recommend:
1. Create state snapshot
2. Run /pause
3. Continue in fresh session

Integration

This skill integrates with:

• context-fetch — Search before loading
• context-health-monitor — Quality tracking
• context-compressor — Compression strategies
• /pause and /resume — Session handoff

Anti-Patterns

❌ Loading files "for context" — Search first ❌ Re-reading same file — Summarize once ❌ Full file when snippet suffices — Target load ❌ Ignoring budget warnings — Quality will degrade

Part of GSD v1.6 Token Optimization. See PROJECT_RULES.md for efficiency rules.