Artifacts Usage Report - May 2026

[github-actions[bot]]

Summary Statistics:

• Total Artifacts: 4,757
• Total Storage: 12.75 GB
• Unique Workflows: 810+ workflow runs
• Analysis Period: 30 days
• Retention: Default (1 day for artifacts)

---

📊 Storage by Artifact Type

Artifact Type	Count	Total Size	Avg Size	Max Size	Workflow Runs
GitHub Pages Build	30	7.13 GB	243.5 MB	243.8 MB	30
Agent Execution Logs	441	2.01 GB	4.7 MB	238.2 MB	439
gh-aw-linux-amd64	157	1.56 GB	10.2 MB	10.3 MB	157
Workflow Activation Artifacts	281	1.55 GB	5.6 MB	5.7 MB	281
Test Results	2,318	219 MB	92 KB	2.0 MB	204
release-binaries-v0.75.3	1	121 MB	121.3 MB	121.3 MB	1
Cache Memory Storage	112	100 MB	911 KB	34.2 MB	112
Code Coverage Reports	29	55 MB	1.9 MB	1.9 MB	29
Repository Memory	55	4 MB	82 KB	1.4 MB	55
safe-outputs-assets	6	4 MB	686 KB	1.6 MB	6
Test Coverage Analysis	29	3 MB	102 KB	102 KB	29
sbom-artifacts	3	2 MB	543 KB	543 KB	3
Threat Detection Reports	208	1 MB	10 KB	31 KB	208
Docker Build Artifacts	48	1 MB	31 KB	51 KB	45
Fuzz Testing Results	317	1 MB	0 KB	10 KB	80
sbom.spdx.json	3	1 MB	358 KB	358 KB	3
sbom.cdx.json	3	1 MB	184 KB	184 KB	3
spellcheck-results	2	0 MB	154 KB	154 KB	2
Benchmark Results	59	0 MB	0 KB	0 KB	59
safe-outputs-conformance-report	131	0 MB	0 KB	0 KB	131

View All Artifact Types (45 total)

• license-report: 89 artifacts, 0.1 MB (89 workflow runs)
• dependency-audit: 80 artifacts, 0.1 MB (80 workflow runs)
• Safe Output Items: 277 artifacts, 0.1 MB (277 workflow runs)
• Secret Validation Reports: 28 artifacts, 0.0 MB (28 workflow runs)
• geo-optimizer-results: 2 artifacts, 0.0 MB (2 workflow runs)
• openai-models: 3 artifacts, 0.0 MB (3 workflow runs)
• gemini-models: 3 artifacts, 0.0 MB (3 workflow runs)
• smokecopilot-experiment: 5 artifacts, 0.0 MB (5 workflow runs)
• smokegemini-experiment: 5 artifacts, 0.0 MB (5 workflow runs)
• anthropic-models: 3 artifacts, 0.0 MB (3 workflow runs)
• smokepi-experiment: 5 artifacts, 0.0 MB (5 workflow runs)
• dailynews-experiment: 3 artifacts, 0.0 MB (3 workflow runs)
• copilot-billing-multipliers: 3 artifacts, 0.0 MB (3 workflow runs)
• dailyastrostylelitemarkdownspellcheck-experiment: 2 artifacts, 0.0 MB (2 workflow runs)
• dailycommunityattribution-experiment: 2 artifacts, 0.0 MB (2 workflow runs)
• dailysecurityredteam-experiment: 2 artifacts, 0.0 MB (2 workflow runs)
• dailysemgrepscan-experiment: 2 artifacts, 0.0 MB (2 workflow runs)
• dailycodemetrics-experiment: 2 artifacts, 0.0 MB (2 workflow runs)
• dailycompilerquality-experiment: 2 artifacts, 0.0 MB (2 workflow runs)
• dailyissuesreport-experiment: 1 artifacts, 0.0 MB (1 workflow runs)
• install-test-windows-latest: 1 artifacts, 0.0 MB (1 workflow runs)
• install-test-ubuntu-latest: 1 artifacts, 0.0 MB (1 workflow runs)
• install-test-ubuntu-22.04: 1 artifacts, 0.0 MB (1 workflow runs)
• install-test-macos-latest: 1 artifacts, 0.0 MB (1 workflow runs)
• code-scanning-sarif: 1 artifacts, 0.0 MB (1 workflow runs)

---

🔍 Key Insights

Largest Storage Consumers

1. GitHub Pages Build: 7.13 GB (56.0% of total)

   • 30 artifacts across 30 workflow runs
   • Average size: 243.5 MB per artifact

2. Agent Execution Logs: 2.01 GB (15.7% of total)

   • 441 artifacts across 439 workflow runs
   • Average size: 4.7 MB per artifact

3. gh-aw-linux-amd64: 1.56 GB (12.3% of total)

   • 157 artifacts across 157 workflow runs
   • Average size: 10.2 MB per artifact

Artifact Distribution Patterns

• Workflow Activation Logs: 281 artifacts (essential for debugging workflow triggers)
• Agent Execution Logs: 441 artifacts (agent session recordings and outputs)
• Test Results: 2318+ artifacts (comprehensive test coverage data)
• Threat Detection: 208+ artifacts (security analysis results)

Storage Trends

• GitHub Pages: Single largest artifact type (7.13 GB for 30 builds) - consider optimizing page assets or reducing build frequency
• CI Build Artifacts: gh-aw binaries account for 1.56 GB - these are essential for distribution but could be pruned more aggressively
• Test Artifacts: Over 2,300 test result artifacts totaling ~214 MB - good coverage with reasonable storage impact
• Activation Logs: ~6 MB per activation × 281 = 1.55 GB - consider compressing or sampling

---

💡 Recommendations

Immediate Actions

1. Optimize GitHub Pages Artifacts (Saves ~7 GB)

   • Pages artifacts are the largest consumer at 7.1 GB
   • Review asset compression and bundling strategies
   • Consider reducing documentation build frequency or using incremental builds

2. Review Binary Retention (Saves ~1.5 GB)

   • gh-aw Linux binaries account for 1.6 GB across 157 artifacts
   • Implement more aggressive pruning (keep only latest + releases)
   • Consider using external storage for release binaries

3. Compress Large Artifacts (Saves ~2 GB)

   • Agent execution logs (2 GB, 441 artifacts) could benefit from compression
   • Average agent log is 4.6 MB - compression could reduce by 60-70%

Long-term Optimizations

4. Implement Tiered Retention

   • Critical artifacts (releases, coverage): 90 days
   • Diagnostic artifacts (agent logs, detection): 7 days
   • Test results: 3 days
   • Activation logs: 1 day (current default is appropriate)

5. Artifact Deduplication

   • Test results show high volume (2,318 artifacts)
   • Consider consolidating per-test artifacts into single archive
   • Implement content-based deduplication for similar test runs

6. Monitoring and Alerts

   • Set up alerts when artifact storage exceeds 15 GB
   • Weekly reports on artifact growth trends
   • Automated cleanup of orphaned artifacts

Expected Impact

Implementing all recommendations could reduce storage by:

• Immediate: 8-10 GB (60-70% reduction)
• Long-term: Maintain < 5 GB steady state
• Cost savings: $0.08/GB/month = ~$0.80/month current, $0.40/month optimized

---

📈 Artifact Lifecycle

Based on the analysis of 4,757 artifacts from 810 workflow runs:

• Creation rate: ~158 artifacts/day (average over 30 days)
• Storage growth: ~425 MB/day
• Peak usage: GitHub Pages builds (238 MB per artifact)
• Most frequent: Test results (77 artifacts/day average)

---

🔬 Sample Workflow Analysis

Based on 18 recently active workflows:

• CGO: 10 artifacts, 14.1 MB across 1 runs
  • Artifact types: test-coverage-analysis-cgo (1), fuzz-results-Workflow-Triggers (1), fuzz-results-Workflow-Core (1)
• PR Sous Chef: 5 artifacts, 11.5 MB across 2 runs
  • Artifact types: activation (2), safe-outputs-items (1), detection (1)
• Safe Output Health Monitor: 6 artifacts, 6.4 MB across 1 runs
  • Artifact types: safe-outputs-items (1), detection (1), githubgh-awRWZCXM.dockerbuild (1)
• CLI Version Checker: 5 artifacts, 6.4 MB across 1 runs
  • Artifact types: safe-outputs-items (1), detection (1), agent (1)
• Static Analysis Report: 6 artifacts, 6.2 MB across 1 runs
  • Artifact types: safe-outputs-items (1), detection (1), githubgh-awX7JIS3.dockerbuild (1)
• Chaos PR Bundle Fuzzer: 5 artifacts, 6.2 MB across 1 runs
  • Artifact types: safe-outputs-items (1), detection (1), agent (1)
• Duplicate Code Detector: 2 artifacts, 6.2 MB across 1 runs
  • Artifact types: agent (1), activation (1)
• Daily CLI Tools Exploratory Tester: 3 artifacts, 6.2 MB across 1 runs
  • Artifact types: githubgh-awK1CGUA.dockerbuild (1), agent (1), activation (1)
• Daily Semgrep Scan: 5 artifacts, 6.1 MB across 1 runs
  • Artifact types: safe-outputs-items (1), detection (1), agent (1)
• GitHub Remote MCP Authentication Test: 4 artifacts, 6.1 MB across 1 runs
  • Artifact types: safe-outputs-items (1), detection (1), agent (1)

---

📋 Methodology

Data Collection:

• Analyzed all non-expired artifacts from the last 30 days
• Queried GitHub Actions API for artifact metadata
• Cross-referenced with workflow run information

Limitations:

• ~98% of artifacts identified and categorized by type
• Some older workflow runs may not have complete metadata
• Artifact retention policies may affect historical data

Tools Used:

• GitHub CLI (gh) for API access
• Python for data aggregation and analysis
• jq for JSON processing

---

Report Generated: 2026-05-24 06:51 UTC
Workflow Run: §26354336826

Generated by 📦 Artifacts Summary · ● son45 1.2M · ◷

• expires on May 25, 2026, 6:54 AM UTC

[github-actions[bot]]

This discussion was automatically closed because it expired on 2026-05-25T06:54:09.101Z.

Closed by Workflow