🏥 Repository Health Report — github/gh-aw (June 2026)

[github-actions[bot]]

This report covers activity in github/gh-aw (GitHub Agentic Workflows), analyzing issues and pull requests from the past 30 days (May 2 – June 1, 2026). The repository is a Go-based GitHub CLI extension that enables writing agentic workflows in natural language as Markdown files, executed as GitHub Actions.

The repository shows exceptional velocity: ~164 issues and ~83 PRs opened per day on average over the last 30 days. Most of this activity is automated (agentic workflows, bots, Copilot agent), reflecting the "dogfooding" nature of gh-aw — it uses its own system heavily.

📊 Volume Metrics

Overall Status (Last 30 Days Sample)

Metric	Count
Issues Opened (30d)	4,897
Issues Closed (30d)	4,829
Issues Currently Open	75
PRs Opened (30d)	2,485
PRs Closed (30d)	2,476
PRs Merged (30d)	1,701
PRs Currently Open	11
Repository Stars	4,549
Repository Forks	412

Note: Data collected from the github/gh-aw repository (the only repo accessible via this workflow's token scope). The repo is extremely active due to agentic automation workflows running continuously.

Recent Activity — 7 Days (May 25 – June 1)

Metric	Count	Daily Avg
Issues Opened	1,136	~162
Issues Closed	1,128	~161
PRs Opened	609	~87
PRs Closed	619	~88

30-Day Trend

🏆 Top Contributors (Last 30 Days)

Contributor	Issues	PRs	Type
github-actions[bot]	3,122	786	🤖 Automation
Copilot	1,608	1,615	🤖 AI Agent
mnkiefer	50	42	👤 Human
dependabot[bot]	38	38	🤖 Automation
dsyme	26	11	👤 Human
lpcox	38	2	👤 Human
szabta89	11	3	👤 Human
salmanmkc	6	6	👤 Human
pelikhan	5	5	👤 Human
mrjf	6	0	👤 Human

Automation accounts for ~93% of all activity, with github-actions[bot] and Copilot agent together opening ~4,730 issues and ~2,401 PRs in 30 days. This is by design — gh-aw runs many agentic workflows against itself (testing, monitoring, daily reports, etc.).

👤 Human Contributors Detail

The top human contributors (mnkiefer, dsyme, lpcox, szabta89, salmanmkc, pelikhan) represent the core maintainer team. Their activity is focused on actual feature development, bug fixes, and code review — contrasting with the high-volume automated issues and PRs from agentic workflows.

🏷️ Label Distribution

Label Counts Table

Label	Count
automation	1,358
agentic-workflows	1,073
cookie	655
testing	400
test-in-progress	395
closed:unknown	266
documentation	222
observability	166
improvement	165
quick-win	137

🔥 High-Activity Open Issues

Issue	Comments	Link
[aw] No-Op Runs	25	#36156
[aw] Detection Runs	3	#35942
[step-names] Align "Start MCP server" step name	2	#36062
[aw-failures] Squid firewall container unhealthy	2	#35780
[CGO] Workflow failure on main	1	#35028

🔄 Open Pull Requests (11)

View All Open PRs

Currently 11 open PRs, mostly from the schema-coverage feature branch adding schema coverage demos. Recent releases show rapid iteration (v0.77.3 → v0.77.4 → v0.77.5 within 3 days), indicating fast merge cycles.

PR Metric	Value
Currently Open	11
Avg Time to Merge	~1.9 hours
Median Time to Merge	~0.8 hours
30d Merge Rate	67.8%

The fast merge cadence (median 48 minutes) reflects an efficient CI/CD pipeline and strong code review culture.

⚠️ Items Needing Attention

Category	Count	Action
Open Issues (total)	75	Monitor for automation noise vs. real bugs
Unassigned Open Issues	60	Triage and assign to appropriate team members
Unlabeled Open Issues	2	Add labels for discoverability
Stale Issues (30d+ old, inactive)	0	✅ None — great hygiene!
Stale PRs (30d+ old, inactive)	0	✅ None — great hygiene!

💡 Commentary and Recommendations

Strengths:

• ✅ Excellent throughput: ~164 issues/day with a near 1:1 open/close ratio indicates the team (and automation) is keeping up with incoming work
• ✅ Fast PR cycle: Median 48-minute merge time is exceptional — fast feedback loops enable rapid iteration
• ✅ No stale work: Zero stale issues or PRs means nothing is falling through the cracks
• ✅ Regular releases: 3 releases in 3 days (v0.77.3–v0.77.5) shows active development

Areas for Attention:

1. 60 unassigned open issues — while many may be automation-generated issues (which auto-close), a periodic triage pass would ensure real bugs get owners
2. no-op runs issue ([aw] No-Op Runs #36156) with 25 comments is the highest-engagement open issue — likely needs a decision on resolution approach
3. Squid firewall container issue ([aw-failures] Squid firewall container (awf-squid) unhealthy → claude engine fails to start (0-turn run failures) #35780) — infrastructure reliability issue worth prioritizing for workflow stability
4. Automation vs. signal ratio — With 93% automated activity, consider adding triage automation to auto-label/assign issues from known workflows to reduce manual triage burden

📋 Data Collection Methodology

Data Sources

• Repository: github/gh-aw (the only repository accessible with the workflow's token scope)
• Issues collected: 4,916 issues from May 2 – June 1, 2026 (pages 1–51 of the issues API)
• PRs collected: 2,510 PRs from May 2 – June 1, 2026 (pages 1–26 of the pulls API)
• Analysis: Python/pandas for data processing, Matplotlib for visualizations

Limitations

• Data represents the last ~51 pages × 100 items = ~5,100 most recent issues; total historical count is higher
• The GitHub search API returned limited cross-org results with the current token scope
• "Open" counts in this dataset reflect items still open from the 30-day window, not the total repository backlog

Workflow Run

§26748261367

Generated by 📊 Organization Health Report · sonnet46 4.3M · ◷

• expires on Jun 2, 2026, 10:22 AM UTC

[github-actions[bot]]

This discussion was automatically closed because it expired on 2026-06-02T10:22:28.902Z.

Closed by Workflow