track-kpis

Track KPIs

Calculate and update agentic coding KPIs to measure ZTE progression.

Arguments

• $ARGUMENTS: State context containing ADW ID, issue info, and workflow history

Instructions

You are updating the agentic KPI tracking file to measure workflow effectiveness.

Step 1: Parse State

Extract from arguments or current context:

• adw_id: Workflow identifier
• issue_number: GitHub issue number
• issue_class: Classification (/​chore, /​bug, /​feature)
• plan_file: Path to implementation plan
• all_adws: List of workflows run (for attempts calculation)

Step 2: Calculate Attempts

Count only workflows that represent restarts:

attempts_incrementing = ["adw_plan_iso", "adw_patch_iso", "plan", "patch"]
attempts = count(workflow for workflow in all_adws if any(inc in workflow for inc in attempts_incrementing))

Build, test, review don't count - only full replans.

Step 3: Get Plan Size

wc -l {plan_file}

Extract line count as plan_size.

Step 4: Get Diff Statistics

git diff origin/​main --shortstat

Parse output to extract:

• Files changed
• Lines added (+)
• Lines removed (-)

Step 5: Update Detail Table

Add new row to the KPI detail table:

Step 6: Recalculate Summary

Current Streak: Count consecutive rows from bottom where Attempts <= 2

Longest Streak: Find longest consecutive sequence where Attempts <= 2

Average Presence: Mean of all attempts values

Total Plan Size: Sum of all plan_size values

Total Diff Size: Sum of (added + removed) across all runs

Step 7: Update KPI File

Write updated summary and detail tables to the KPI file (typically app_docs/​agentic_kpis.md or project-specific location).

Output

Report KPI update:

{
  "success": true,
  "this_run": {
    "adw_id": "{adw_id}",
    "issue": "{issue_number}",
    "attempts": 1,
    "plan_size": 45,
    "diff_added": 67,
    "diff_removed": 23,
    "files_changed": 4
  },
  "summary": {
    "current_streak": 6,
    "longest_streak": 12,
    "average_presence": 1.28,
    "total_plan_size": 450,
    "total_diff_size": 2340
  }
}

Notes

• Lower attempts = better (target: 1)
• Higher streak = better (demonstrates consistency)
• Track trends over time to assess ZTE readiness
• 90%+ success rate (attempts <= 2) indicates ZTE readiness