📊 Daily Copilot Token Consumption Report - 2025-12-04

[github-actions[bot]]

📊 Daily Copilot Token Consumption Report - 2025-12-04

Executive Summary

Over the last 30 days, Copilot-powered agentic workflows consumed 488,046 tokens at an estimated cost of $7.32, across 24 workflow runs covering 11 unique workflows.

Key Highlights:

• Highest consuming workflow: Running Copilot coding agent (341,504 tokens, $5.12)
• Most active workflow: Running Copilot coding agent (8 runs)
• Average cost per run: $0.305
• Trend: Token usage is decreasing by 82.3% compared to previous measurement

Full Report Details

📈 Token Usage Trends

Overall Trends

Due to environment limitations, visual charts could not be generated. Below is the tabular data:

Daily Token Consumption:

Date	Total Tokens	Total Cost	Total Runs
2025-12-04	2,762,746	$41.44	406
2025-12-04	488,046	$7.32	24

Trend Analysis: Token consumption has decreasing by 82.3% between measurements. The most recent measurement shows 488,046 tokens consumed across 24 workflow runs, compared to the earlier measurement of 2,762,746 tokens across 406 runs.

🏆 Top Workflows by Token Consumption

Top 10 Workflows

Rank	Workflow	Total Tokens	Total Cost	Runs	Avg Tokens/Run	Avg Cost/Run
1	Running Copilot coding agent	341,504	$5.12	8	42,688	$0.640
2	Addressing comment on PR #5486	40,000	$0.60	4	10,000	$0.150
3	Addressing comment on PR #5487	40,000	$0.60	4	10,000	$0.150
4	GitHub MCP Structural Analysis	10,000	$0.15	1	10,000	$0.150
5	Copilot PR Conversation NLP Analysis	10,000	$0.15	1	10,000	$0.150
6	Copilot PR Prompt Pattern Analysis	10,000	$0.15	1	10,000	$0.150
7	Static Analysis Report	10,000	$0.15	1	10,000	$0.150
8	Daily Copilot Token Consumption Report	7,666	$0.11	1	7,666	$0.115
9	Smoke Copilot Playwright	6,566	$0.10	1	6,566	$0.098
10	Smoke Copilot No Firewall	6,311	$0.09	1	6,311	$0.095

📊 All Workflow Statistics

View All 11 Workflows

Workflow	Total Tokens	Total Cost	Runs	Avg Tokens	Avg Cost	Avg Turns	Avg Duration
Running Copilot coding agent	341,504	$5.12	8	42,688	$0.640	5	N/A
Addressing comment on PR #5486	40,000	$0.60	4	10,000	$0.150	5	N/A
Addressing comment on PR #5487	40,000	$0.60	4	10,000	$0.150	5	N/A
GitHub MCP Structural Analysis	10,000	$0.15	1	10,000	$0.150	5	N/A
Copilot PR Conversation NLP Analysis	10,000	$0.15	1	10,000	$0.150	5	N/A
Copilot PR Prompt Pattern Analysis	10,000	$0.15	1	10,000	$0.150	5	N/A
Static Analysis Report	10,000	$0.15	1	10,000	$0.150	5	N/A
Daily Copilot Token Consumption Report	7,666	$0.11	1	7,666	$0.115	5	N/A
Smoke Copilot Playwright	6,566	$0.10	1	6,566	$0.098	5	N/A
Smoke Copilot No Firewall	6,311	$0.09	1	6,311	$0.095	5	N/A
Smoke Copilot	5,999	$0.09	1	5,999	$0.090	5	N/A

💡 Insights & Recommendations

High-Cost Workflows

The following workflows account for the majority of token consumption:

1. Running Copilot coding agent - $5.12 (70.0% of total)

   • Observation: This workflow ran 8 times with an average cost of $0.640 per run
   • Recommendation: Monitor this workflow for optimization opportunities as it runs frequently

2. Addressing comment on PR Add Python script support to safe-inputs with modular MCP handler architecture and dynamic dependency tracking #5486 - $0.60 (8.2% of total)

   • Observation: This workflow ran 4 times with an average cost of $0.150 per run
   • Recommendation: Usage appears reasonable for this workflow type

3. Addressing comment on PR Move setup guide to install.md with init and workflow import options #5487 - $0.60 (8.2% of total)

   • Observation: This workflow ran 4 times with an average cost of $0.150 per run
   • Recommendation: Usage appears reasonable for this workflow type

Optimization Opportunities

1. High-Token Workflows: 0 workflows have average token usage above 50,000 tokens per run

   • Affected Workflows:
   • Potential Savings: ~$0.00 per month (20% reduction estimate)
   • Action: Review prompt complexity and context size for these workflows

2. Frequent Runners: 3 workflows run more than 3 times in this period

   • Affected Workflows: Running Copilot coding agent, Addressing comment on PR Add Python script support to safe-inputs with modular MCP handler architecture and dynamic dependency tracking #5486, Addressing comment on PR Move setup guide to install.md with init and workflow import options #5487
   • Action: Consider caching common responses or reducing trigger frequency

Efficiency Trends

• Token efficiency: Average of 20,335 tokens per run
• Cost efficiency: Average cost per run is $0.305
• Run patterns: Most activity from "Running Copilot coding agent" with 8 runs

📅 Historical Comparison

Metric	Current Period	Previous Period	Change
Total Tokens	488,046	2,762,746	-82.3%
Total Cost	$7.32	$41.44	-82.3%
Total Runs	24	406	-94.1%
Avg Cost/Run	$0.305	$0.102	+198.8%

🔧 Methodology

• Data Source: GitHub Actions workflow run API (last 30 days)
• Engine Filter: Copilot engine only
• Cache Storage: /tmp/gh-aw/cache-memory/token-metrics/
• Analysis Date: 2025-12-04T11:09:27.962977
• Historical Data: 2 measurement periods
• Cost Model: Based on Copilot token pricing ($0.000015 per token estimate)

📊 Data Quality Notes

• Token usage estimates are based on historical averages where artifact data is unavailable
• Some workflows may not have complete cost data if artifacts were not preserved
• The report covers the most recent workflow runs that match the Copilot engine filter
• Trend analysis is based on comparing sequential measurement periods

---

Generated by Daily Copilot Token Consumption Report
Next report: Tomorrow at 11 AM UTC (weekdays only)

AI generated by Daily Copilot Token Consumption Report

[github-actions[bot]]

This discussion was automatically closed because it was created by an agentic workflow more than 3 days ago.