📊 Copilot Token Usage Report2026-05-28

[github-actions]

Overview

Period: 2026-05-27 to 2026-05-28
Runs analyzed: 41 (39 had token data)
Total tokens: 17,851,043 (~17.9M) across all workflows
Estimated total cost: $0.00 (cost field not populated — api-proxy may not report USD costs for Copilot engine)
Total Actions minutes: 210 min
Total turns: 423

⚠️ estimated_cost is $0 for all runs — Copilot API usage may be billed differently. Token counts are the primary efficiency metric here.

Workflow Summary

Workflow	Runs	Total Tokens	Avg Tokens/Run	Turns	Avg Turns/Run
Test Coverage Improver	1	3,768K	3,768K	61	61
Duplicate Code Detector	1	2,167K	2,167K	41	41
Build Test Suite	3	1,280K	427K	33	11
API Surface & Export Audit	1	1,205K	1,205K	27	27
Smoke Services	5	1,328K	266K	42	8.4
Firewall Issue Dispatcher	4	1,153K	288K	38	9.5
Daily Copilot Token Optimization Advisor	1	805K	805K	18	18
Daily Security Review and Threat Modeling	1	718K	718K	15	15
CI/CD Pipelines and Integration Tests Gap Assessment	1	921K	921K	17	17
Schema & Spec Sync	1	594K	594K	18	18
Refactoring Opportunity Scanner	1	580K	580K	13	13
Test Coverage Reporter	2	620K	310K	17	8.5
Pelis Agent Factory Advisor	1	391K	391K	10	10
Smoke Copilot	5	756K	151K	24	4.8
Smoke Copilot BYOK	5	607K	121K	21	4.2
Daily Claude Token Usage Analyzer	1	269K	269K	7	7
Issue Duplication Detector	1	267K	267K	7	7
Dependency Security Monitor	1	185K	185K	5	5
Smoke Chroot	3	237K	79K	9	3
Daily Copilot Token Usage Analyzer	1	0	—	0	—
Daily Claude Token Optimization Advisor	1	0	—	0	—

🔍 Optimization Opportunities

1. Test Coverage Improver — 3,768K tokens, 61 turns in a single run

   • Extremely high turn count (4× the >15 turn threshold). Consider breaking into smaller scoped tasks or pre-computing which files need coverage before the agent run.
   • Review if the agent is doing repetitive read/search cycles that could be pre-filtered.

2. Duplicate Code Detector — 2,167K tokens, 41 turns in a single run

   • High turns suggest iterative scanning. Consider pre-computing candidate file pairs with static analysis tools and passing results to the agent.

3. API Surface & Export Audit — 1,205K tokens, 27 turns

   • Reduce by pre-extracting exported symbols with ts-morph or tsc --declaration before the agent run.

4. Build Test Suite — 1,280K tokens across 3 runs, avg 427K tokens / 11 turns per run

   • Consistently high. Consider whether test generation context (full source) can be narrowed to changed files only.

5. Smoke Services — 1,328K total tokens across 5 runs, avg 266K / 8.4 turns

   • Highest cumulative token usage among recurring smoke tests. Review if system prompt or tool surface can be reduced.

Per-Workflow Details

Test Coverage Improver

• Runs: 1 (§26566078232 — success)
• Total tokens: 3,768K
• Turns: 61
• Error rate: 0/1 runs

Duplicate Code Detector

• Runs: 1 (§26541863694 — success)
• Total tokens: 2,167K
• Turns: 41
• Error rate: 0/1 runs

Build Test Suite

• Runs: 3 (§26520287227, §26519635883, §26519419114 — all success)
• Total tokens: 1,280K (avg 427K/run)
• Turns: 33 total (avg 11/run)
• Error rate: 0/3 runs

Smoke Services

• Runs: 5 (all success)
  • §26548402795, §26520287287, §26519635964, §26519418946, §26514353803
• Total tokens: 1,328K (avg 266K/run)
• Turns: 42 total (avg 8.4/run)
• Error rate: 0/5 runs

Firewall Issue Dispatcher

• Runs: 4 (all success)
  • §26561631618, §26548559344, §26532725455, §26514560273
• Total tokens: 1,153K (avg 288K/run)
• Turns: 38 total (avg 9.5/run)
• Error rate: 0/4 runs

CI/CD Pipelines and Integration Tests Gap Assessment

• Runs: 1 (§26557571025 — success)
• Total tokens: 921K
• Turns: 17 (at the >15 threshold)
• Error rate: 0/1 runs

API Surface & Export Audit

• Runs: 1 (§26531511236 — success)
• Total tokens: 1,205K
• Turns: 27
• Error rate: 0/1 runs

Daily Copilot Token Optimization Advisor

• Runs: 1 (§26503705843 — success)
• Total tokens: 805K
• Turns: 18
• Error rate: 0/1 runs

Daily Security Review and Threat Modeling

• Runs: 1 (§26514335886 — success)
• Total tokens: 718K
• Turns: 15
• Error rate: 0/1 runs

Schema & Spec Sync

• Runs: 1 (§26514585106 — success)
• Total tokens: 594K
• Turns: 18
• Error rate: 0/1 runs

Refactoring Opportunity Scanner

• Runs: 1 (§26523529925 — success)
• Total tokens: 580K
• Turns: 13
• Error rate: 0/1 runs

Test Coverage Reporter

• Runs: 2 (§26557413797, §26531511051 — both success)
• Total tokens: 620K (avg 310K/run)
• Turns: 17 total (avg 8.5/run)
• Error rate: 0/2 runs

Smoke Copilot

• Runs: 5 (all success) — §26548610617, §26520287286, §26519635882, §26519418981, §26514735172
• Total tokens: 756K (avg 151K/run)
• Turns: 24 total (avg 4.8/run)

Smoke Copilot BYOK

• Runs: 5 (all success) — §26548614680, §26520287157, §26519635971, §26519418939, §26514757323
• Total tokens: 607K (avg 121K/run)
• Turns: 21 total (avg 4.2/run)

Workflows Without Token Data

Runs where token_usage was null or 0 — these may not have the api-proxy enabled or are still in-progress:

• Daily Copilot Token Usage Analyzer — §26566994791 (current run, in progress)
• Daily Claude Token Optimization Advisor — §26566872974

Historical Trend

This analysis covers a 24-hour window (2026-05-27 to 2026-05-28). No previous Copilot token usage report issues were found to compare against — this may be the first report, or previous reports used different labels.

References: §26566994791

Generated by Daily Copilot Token Usage Analyzer · sonnet46 1M · ◷

[github-actions]

This issue is being closed as outdated. A newer issue has been created: #3993

View newer issue

---

This action was performed automatically by the Daily Copilot Token Usage Analyzer workflow.