Impact Efficiency Report - 2026-06-16

[github-actions]

Analysis window: 2025-12-18 → 2026-06-16 (180 days) · Run 27633305387

Executive Summary

Across 180 days, agentic workflows delivered 3 accepted PR outcomes (all merged) with a total outcome value of 205 against 43,055 AIC total spend. The highest-impact work was two bug fixes in the safe-outputs stack (PRs #37863, #37565, bug label, 85 value each), plus one new Go linter (PR #38185, improvement, 35 value). Repo-wide Impact Efficiency is 0.00476 value/AIC — low because 993 of 1,000 analyzed PRs were excluded for lacking a Closes #N link, not because valuable work wasn't done. Top AIC consumers (daily-experiment-report 3,625 AIC; Contribution Check 2,278 AIC) produced zero PR outcomes in scope.

Summary

Metric	Value
PRs analyzed with linked issue	7 (merged: 3, closed: 4)
PRs excluded (no linked closing issue)	993 (merged: 497, closed: 496)
Safe output outcomes analyzed	0
Outcomes mapped to objectives	7
Unmapped outcomes	0
Accepted outcome count	3
Total outcome value	205
AI Credits (all workflows)	43,055.27
Impact Efficiency	0.00476 value/AIC

Agentic Work by Objective

🐛 Bug — value: 85 · 2 outcomes · total value: 170

Per-outcome AIC (PR body data, secondary): 296.4 + 190.0 = 486.4 AIC · Category IE: 0.350

• PR #37863 → root #37835 (bug, compiler) · Fix push_to_pull_request_branch head ref
• PR #37565 → root #37545 (bug, compiler) · Fix multi-repo checkout after git-credentials clobber

🔧 Improvement — value: 35 · 1 outcome · total value: 35

Per-outcome AIC (PR body data, secondary): 389.9 AIC · Category IE: 0.090

• PR #38185 → root #38143 (improvement, quick-win) · [linter-miner] add execcommandwithoutcontext analyzer

⚠️ Highest AIC-consuming category: Bug. Also highest-value; spend was justified. Per-category AIC uses PR-body run costs (secondary source), not full producing-workflow costs.

Which Workflows Drove That Impact

Attribution source: <!-- gh-aw-agentic-workflow: ... --> HTML comments in PR bodies. workflow-logs.json is empty — no deterministic run attribution available.

PR Description Updater (pr-description-caveman)

Metric	Value
Accepted, mapped outcomes	3
Total outcome value	205 (100% share)
Workflow AIC (7 runs, aic-by-workflow.json)	633.26
Workflow Impact Efficiency	0.324 value/AIC

Top categories: Bug (170, 2 outcomes), Improvement (35, 1 outcome)

⚠️ PR Description Updater enriches PR descriptions; it may not be the PR-creating workflow. PR #38185 title prefix [linter-miner] suggests a different creator (not in AIC data). Attribution is best-effort from HTML comments.

Highest-value workflow: PR Description Updater (205, 100%) · Highest AIC workflow: daily-experiment-report (3,625 AIC, 0 attributed outcomes)

Top 10 AIC workflows with zero attributed outcomes

Workflow	AIC	Runs
daily-experiment-report	3,625	2
Impact Efficiency Report	2,647	3
Contribution Check	2,278	5
Daily Semgrep Scan	2,209	2
Daily Firewall Logs Collector	2,144	1
Glossary Maintainer	2,039	2
Daily CLI Tools Exploratory Tester	1,792	2
Package Specification Extractor	1,767	2
Documentation Noob Tester	1,285	2
Terminal Stylist	1,000	2

These 10 consumed 20,787 AIC (48.3%) with 0 in-scope outcome value — likely legitimate analytical/security work outside the PR+linked-issue model.

Top Outcomes by Outcome Value

Outcome	Workflow	Type	Root Issue	Obj. Value	Outcome Value
PR #37863 Fix #37835: push_to_pull_request_branch	PR Description Updater	PR merged	#37835 bug	85	85
PR #37565 Fix #37545: multi-repo checkout	PR Description Updater	PR merged	#37545 bug	85	85
PR #38185 [linter-miner] add linter 1	PR Description Updater	PR merged	#38143 improvement	35	35
PR #36767 [docs] healer 2026-06-04	Daily Documentation Healer	PR closed	#36275 improvement	35	0
PR #36371 [docs] healer 2026-06-02	Daily Documentation Healer	PR closed	#34686 improvement	35	0
PR #35906 [docs] updater 2026-05-30	Daily Documentation Updater	PR closed	#35506 improvement	35	0
PR #35330 [docs] healer 2026-05-28	Daily Documentation Healer	PR closed	#34686 improvement	35	0

1 PR #38185 [linter-miner] title prefix suggests a different creating workflow; HTML comment says PR Description Updater.

Unmapped Outcomes

None. All 7 in-scope outcomes mapped via improvement or bug labels (max-value logic; cookie/deep-report labels are not in the mapping but co-occur with mapped labels).

Interpretation

Accepted count (3) is easy to game — closing cosmetic issues inflates it. Impact Efficiency (0.00476 value/AIC) better reflects cost-adjusted value, but the denominator includes all 78 workflows while only 7 PRs qualified. The dominant bottleneck is low Closes #N linkage (0.7% of PRs), not insufficient agentic work.

• Highest value/AIC category: Bug (0.35 value/AIC) — the two safe-outputs stack fixes were substantive
• Most efficient workflow: PR Description Updater (0.324 value/AIC; attribution uncertain)
• Highest AIC with no attributed value: daily-experiment-report (3,625 AIC) — likely outside current scope, not necessarily wasted

Data Quality

Dimension	Status
PR root-issue tracing	⚠️ 993/1000 excluded — only 0.7% of PRs had Closes #N
Root-issue label fetches	✅ Live gh issue view for 6 issues — all succeeded
Safe-output outcomes	i️ 0 — evaluations file is empty
Workflow attribution	⚠️ workflow-logs.json empty — HTML comment proxy only
AIC source	✅ Deterministic aic-by-workflow.json (48 snapshots, token-audit-memory)
Per-run AIC (per-category)	⚠️ Secondary — PR body footers (876.3 AIC sum vs. 633.26 in aic-by-workflow.json)
linter-miner AIC	⚠️ Unknown — no entry in aic-by-workflow.json

Human Work

workflow-logs.json is empty → all 500 merged PRs technically have "no matching run." Using HTML comment proxies: 3 merged PRs with linked issues all carry agentic HTML comments (counted in agentic analysis above). 0 confirmed human-authored merged PRs with linked closing issues identified.

Metric	Value
Total merged PRs	500
With agentic HTML comment (excluded)	3
Possible human-authored	497
With linked closing issue	0
Mapped to objective	0

No objective categories to report for human work. Enable workflow-logs.json data collection and enforce Closes #N in PR templates to unlock this analysis.

---

AIC: aic-by-workflow.json (deterministic) · Root labels: live GitHub API · Run 27633305387

Generated by 📊 Impact Efficiency Report · 641.3 AIC · ⌖ 53.4 AIC · ⊞ 22.8K · ◷

[github-actions]

Superseded by a freshly generated report for the same day (2026-06-16). A new Impact Efficiency Report issue will be created momentarily with updated analysis.