Impact Efficiency Report - 2026-06-16

[github-actions]

Executive Summary

Over the 180-day window (2025-12-18 to 2026-06-16), 3 of 7 qualifying PR outcomes were accepted, delivering 205 total objective value. The highest-impact work was two bug fixes (PR #37863, PR #37565) resolving critical safe-outputs and multi-repo checkout failures, contributing 170 of 205 value. All accepted outcomes are attributed to PR Description Updater (0.32 value/AIC). Repository-wide Impact Efficiency is 0.005 — 43,055 AIC was spent mostly on analytics/reporting workflows with no linked-issue PR outcomes; 993 of 1,000 sampled PRs were excluded for lacking Closes #N.

---

Summary

Metric	Value
PRs analyzed with linked issue	merged: 3, closed: 4, open excluded: 0
PRs excluded (no linked closing issue)	993 (497 merged + 496 closed)
Safe output outcomes analyzed	0
Outcomes mapped to objectives	7
Unmapped outcomes	0
Accepted outcome count	3
Total outcome value	205
AI Credits	43,055.27
Impact Efficiency	0.005 value units / AIC

---

Agentic Work by Objective

bug (value: 85) — Highest-impact

Accepted	Total Value	AIC (approx.)	Efficiency (approx.)
2	170	~181	~0.94

• PR #37863 -> #37835: Fix push_to_pull_request_branch branch derivation
• PR #37565 -> #37545: Fix multi-repo checkout when Configure Git credentials clobbers origin

improvement (value: 35)

Accepted	Total Value	AIC (approx.)	Efficiency (approx.)
1	35	~90	~0.39

• PR #38185 -> #38143: Add execcommandwithoutcontext Go linter analyzer

Highest AIC category (no value): Analytics/reporting stack consumed ~18,502 AIC (43% of total) — daily-experiment-report (3,625), Impact Efficiency Report (2,647), Contribution Check (2,278), Daily Semgrep Scan (2,209), Daily Firewall Logs (2,144) — with zero linked-issue PR outcomes.

---

Which Workflows Drove That Impact

PR Description Updater — Most value (100% share)

Outcomes	Value	AIC	Efficiency
3 accepted	205	633.26	0.32

Top categories: bug (170), improvement (35). Attribution based on gh-aw-agentic-workflow body metadata; creating workflow may differ (workflow-logs.json is empty).

Daily Documentation Healer — 0 value

3 analyzed outcomes (all rejected): PRs #36767, #36371, #35330. AIC: unknown (not in aic-by-workflow.json).

Daily Documentation Updater — 0 value

1 analyzed outcome (rejected): PR #35906. AIC: 323.05.

Top AIC consumers with zero attributed outcomes

Workflow	AIC	Runs
daily-experiment-report	3,625.33	2
Impact Efficiency Report	2,647.42	3
Contribution Check	2,278.16	5
Daily Semgrep Scan	2,208.89	2
Daily Firewall Logs Collector and Reporter	2,143.74	1
Glossary Maintainer	2,039.17	2
Daily CLI Tools Exploratory Tester	1,792.41	2
Package Specification Extractor	1,766.81	2

---

Top Outcomes by Outcome Value

Outcome	Workflow	Type	Root Objective	Obj Value	Outcome Value
PR #37863	PR Description Updater	Merged	#37835 bug	85	85
PR #37565	PR Description Updater	Merged	#37545 bug	85	85
PR #38185	PR Description Updater	Merged	#38143 improvement	35	35
PR #36767	Daily Doc Healer	Closed	#36275 improvement	35	0
PR #36371	Daily Doc Healer	Closed	#34686 improvement	35	0
PR #35906	Daily Doc Updater	Closed	#35506 improvement	35	0
PR #35330	Daily Doc Healer	Closed	#34686 improvement	35	0

---

Unmapped Outcomes

None. All 7 outcomes mapped via objective-mapping.json. Labels cookie, deep-report, linting, go, go-linters, closed:unknown are absent from the mapping and ignored.

---

Interpretation

• Impact Efficiency (0.005) vs accepted count (3): count alone hides importance. Bug fixes (85 pts each) are ~2.4x more valuable than the linter improvement (35 pts).
• Bug category is most cost-efficient (~0.94 value/AIC) — highest-value, lowest estimated cost.
• PR Description Updater is the only workflow with attributed delivered value (0.32 efficiency). Attribution caveat: it is a description-update workflow; the underlying code authors may be different.
• Daily Documentation Healer: 3 closed-without-merge documentation PRs suggest a systematic acceptance issue. AIC unknown.
• Top 8 AIC workflows (43%) have zero linked-issue outcomes; value likely delivered through reports and non-PR channels not covered by this metric.

---

Data Quality

Gap	Notes
PR linked-issue coverage	99.3% excluded (993/1000). Leading metric gap.
Safe-output outcomes	evaluations.jsonl is empty; 0 outcomes.
Workflow attribution	workflow-logs.json is empty; attribution from body markers only. Merged PR creating workflows unknown.
Daily Doc Healer AIC	Not in aic-by-workflow.json; impact efficiency not computable.
Per-category AIC	Approximated: 633.26 total / 7 runs = ~90.5 AIC/run for PR Description Updater.

AIC source: Deterministic — aic-by-workflow.json (48 snapshots, token-audit-memory). Total: 43,055.27. No live fallback used.

---

Human Work

369 merged PRs identified as potentially human-authored (no gh-aw-agentic-workflow body marker). All 3 PRs with linked issues had agentic markers, so zero human PRs are traceable to objectives.

Metric	Value
Potentially human-authored merged PRs	369
With linked closing issue	0
Mapped to objective	0

No AIC computed for human work. Caveat: empty workflow-logs.json means human/agentic distinction relies solely on body markers and may inflate the human count.

---

Impact Efficiency Report workflow · Run #27643083066 · 2026-06-16 · Window: 2025-12-18 to 2026-06-16 (180 days) · Replaces #39612

Generated by 📊 Impact Efficiency Report · ◷