[portfolio] 📊 Portfolio Dashboard - 2026-02-23 #17867
Closed
Replies: 1 comment
-
|
This discussion was automatically closed because it expired on 2026-02-24T09:25:57.138Z.
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Analysis of 313 workflow runs across 80 distinct workflows from Feb 21–23, 2026. Total cost: $104.69 (2-day window). The biggest issue is Issue Monster (66 runs, 100% failure rate) consuming substantial compute with zero value delivered.
Visual Summary
Cost Trends (Last 2 Days)
Key Insights:
Top Cost Drivers
Top 5 workflows account for ~48% of total cost:
Failure Analysis
Critical: 5 workflows with >30% failure rate — 3 at 100% failure
Overall Health
💰 Cost Reduction Opportunities
Estimated Monthly Savings: ~$312–$450/month (20–29% reduction)
Strategy 1: Disable 100%-Failure Workflows — ~$200+/month
Three workflows have never succeeded in the observed window and are burning compute:
Issue Monster alone ran 66 times in 2 days — if it has any AI cost per run, disabling it would be the single highest-impact action. Even at $0.50/run its 2-day compute waste is $33 → ~$500/month.
Fix: In each workflow's
.mdfile, either correct the underlying error or add thedisabled: truefrontmatter field until fixed.Strategy 2: Reduce High-Cost Workflow Frequency — ~$100/month
Several expensive workflows ran 2–3 times in 2 days (triggered by
workflow_dispatchor schedule):If documentation-maintenance workflows (Unbloat, Consolidator, Daily Doc Updater) are set to daily, reducing to weekly would save
3/4 of their cost ≈ **$90/month** combined.Fix (example for
documentation-unbloat.md):Strategy 3: Fix 50%-Failure Workflows — Recover $4–8/month each
Action: Review error logs for each workflow. The
error_countfield in run data will point to the specific step failing.🎯 Priority Actions
CRITICAL — Disable
Issue Monsterimmediately. It ran 66× in 2 days with 100% failure. Even with minimal per-run cost this is the highest-volume waste. Check.github/workflows/issue-monster.mdfor the trigger configuration causing excessive runs.HIGH — Fix or disable
PR Triage Agent(7 runs, 100% failure). Check recent run logs for the root error — likely an authentication or permissions issue.HIGH — Reduce schedule frequency for documentation maintenance workflows (
documentation-unbloat.md,developer-documentation-consolidator.md) from daily → weekly. Saves ~$90/month.MEDIUM — Audit
Go Logger Enhancement(50% failure, $15.92 total) — determine if this is a one-time task that should be removed from schedule after completion.📈 Data Quality
References: §22299460565
Beta Was this translation helpful? Give feedback.
All reactions