Skip to content

πŸ“Š Copilot Token Usage Report2026-05-09Β #2793

@github-actions

Description

@github-actions

Overview

Period: 2026-05-08 to 2026-05-09
Runs analyzed: 50 (48 had token data)
Total tokens: 14,977,275 (~15.0M)
Estimated total cost: $0.00 (cost tracking not configured β€” api-proxy does not report USD cost for this engine)
Total Actions minutes: 216 min
Total turns: 46

Workflow Summary

Workflow Runs Total Tokens Avg Tokens/Run Turns Errors
Build Test Suite 5 4,712K 942K 5 0
API Surface & Export Audit 4 2,006K 502K 4 0
Firewall Issue Dispatcher 2 1,346K 673K 2 0
Duplicate Code Detector 1 593K 593K 1 0
Dependency Security Monitor 1 506K 506K 1 0
CI/CD Gap Assessment 1 381K 381K 1 0
Smoke Services 6 1,484K 247K 6 0
Documentation Maintainer 1 238K 238K 1 0
Issue Duplication Detector 3 544K 181K 3 0
Test Coverage Reporter 5 785K 157K 3 2
Smoke Copilot 8 1,147K 143K 8 0
Smoke Copilot BYOK 7 925K 132K 7 0
Smoke Chroot 4 309K 77K 4 0
Daily Claude Token Analyzer 1 0 0 0 0
Daily Copilot Token Analyzer 1 0 0 0 0

πŸ” Optimization Opportunities

  1. Build Test Suite β€” 942K avg tokens/run (5 runs)

    • Highest per-run token count; exceeds 100K threshold by ~9Γ—
    • Consider whether the agent is analyzing too many files or receiving a large system prompt
    • Significant variance (370K–1,647K) suggests inconsistent workload β€” investigate what differs between runs
  2. Firewall Issue Dispatcher β€” 673K avg tokens/run (2 runs)

    • Very high per-run consumption; also exceeded 100K threshold by ~6Γ—
    • Large run-to-run variance (558K vs 789K) β€” may depend on issue/PR body size
  3. API Surface & Export Audit β€” 502K avg tokens/run (4 runs)

    • High variance (205K–787K) suggests differing input sizes
    • Review if the full codebase is passed as context each run
  4. Duplicate Code Detector β€” 593K tokens in a single run

    • One-shot analysis but very high; consider chunking or scoping to changed files only
  5. Dependency Security Monitor β€” 506K tokens in a single run

    • Similarly high for a single-turn workflow; verify scope of files analyzed
  6. Test Coverage Reporter β€” 40% error rate (2/5 runs failed)

    • Reliability issue β€” investigate root cause of failures
    • Failing runs produced 0 tokens, wasting Actions minutes
Per-Workflow Run Details

Build Test Suite (5 runs)

API Surface & Export Audit (4 runs)

Firewall Issue Dispatcher (2 runs)

Smoke Services (6 runs)

  • [success] 216K–310K tokens (avg 247K) β€” 6 successful runs

Smoke Copilot (8 runs)

  • [success] 114K–234K tokens (avg 143K) β€” 8 successful runs

Smoke Copilot BYOK (7 runs)

  • [success] 84K–237K tokens (avg 132K) β€” 7 successful runs

Test Coverage Reporter (5 runs)

Issue Duplication Detector (3 runs)

  • [success] 160K–193K tokens (avg 181K) β€” 3 successful runs

Duplicate Code Detector (1 run)

Dependency Security Monitor (1 run)

CI/CD Gap Assessment (1 run)

Documentation Maintainer (1 run)

Smoke Chroot (4 runs)

  • [success] 77K–78K tokens (avg 77K) β€” very consistent; 4 successful runs
Workflows Without Token Data

Runs where token_usage was null or 0 β€” these may not have the api-proxy enabled or are currently in-flight:

  • Daily Copilot Token Usage Analyzer β€” Β§25597053980 (this run)
  • Daily Claude Token Usage Analyzer β€” Β§25596878548 (in-flight or no token data)
  • Test Coverage Reporter β€” 2 of 5 runs failed before producing token data

Historical Trend

This appears to be the first Copilot token usage report (no previous token-usage-report issues found). Future reports will include trend comparisons.

References:

Generated by Daily Copilot Token Usage Analyzer Β· ● 314.4K Β· β—·

Metadata

Metadata

Assignees

No one assigned

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions