| name | codeburn |
| description | Use when you want to see token consumption by task type, model, one-shot rate, or cost — observability dashboard for GitHub Copilot sessions. Trigger with /codeburn to analyze session spending. |
| user-invocable | true |
| disable-model-invocation | true |
Codeburn — Token Cost Observability
Use this skill when you want to understand token consumption patterns for your Copilot sessions.
What it does
Reads VS Code session data and provides a breakdown of:
- Token usage by task category (refactor, bug-fix, feature, test, docs, review, debug, config, migration, research, security, cleanup, other)
- Cost by model tier
- One-shot rate (tasks completed in a single turn vs multi-turn)
- USD cost estimate for the session or time period
How to invoke
Type /codeburn in chat, optionally with:
/codeburn today — today's sessions only/codeburn week — last 7 days/codeburn summary — concise cost overview
How Copilot uses this skill
- Reads available session context from VS Code
- Categorizes requests by task type based on conversation history
- Estimates token consumption per category
- Reports top cost drivers and one-shot completion rate
Task categories
| Category | Trigger words | Cost impact |
|---|
feature | add, implement, create, build | High |
refactor | refactor, clean, restructure | High |
bug-fix | fix, bug, error, broken | Medium |
test | test, spec, coverage | Medium |
review | review, audit, check | Medium |
debug | debug, trace, investigate | High |
docs | docs, comment, readme | Low |
research | what, how, explain | Low |
config | config, setup, install | Low |
migration | migrate, upgrade, update | Medium |
security | security, vuln, auth | High |
cleanup | remove, delete, cleanup | Low |
Output format
━━━ Codeburn Report ━━━━━━━━━━━━━━━━━━━
📅 Period: today (2026-01-01)
🔢 Total requests: 47
💰 Estimated cost: $0.82 (Sonnet tier)
Top categories:
feature ████████████ 12 requests (26%)
debug ████████ 8 requests (17%)
refactor ██████ 6 requests (13%)
One-shot rate: 72% (34/47 tasks in single turn)
Multi-turn: 28% (avg 3.2 turns/task)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Cost optimization tips
- High multi-turn rate on
debug → add more context upfront
- Frequent
research requests → consolidate into single queries
- High
refactor cost → scope changes more precisely before starting
