⚡

Developer 3s Key Decision Metrics

TL;DR Verdict
🏆 Benchmark Rank72s 必看
Top Tier Benchmark
Full Leaderboard→
💰 Cost & Plans比价
API & Subscription
Compare Plans↗
🛠️ Ecosystem & IDERules
gh-aw · copilot · claude-code
Setup IDE Rules↗

gh-aw v0.90.1 (2026-09-30T17:20:00Z) adds standalone safe-output-backed ledgers with replay projections (defaults preserved on shared-workflow import), richer audit/usage artifacts (ledger txs, threat-detection outcomes, experiment/evals, friction costs), narrowed add-labels + required-labels gating for assign-to-agent, compile-time self-hosted runner enforcement, kebab-case sandbox frontmatter, and Claude Code CLI pin 2.1.280.

Key Takeaways

  • ✓v0.90.1 at 2026-09-30T17:20:00Z after v0.90.0
  • ✓Standalone safe-output ledgers + replay; defaults kept on shared-workflow import
  • ✓Richer audit: ledger txs, threat detection, evals, friction-cost attribution
  • ✓Governance: narrowed add-labels; required-labels on assign-to-agent; compile-time self-hosted runner enforcement
  • ✓Claude Code CLI pin 2.1.280; docs via gh-aw README / GitHub Next project page
🧭

Finished reading? Explore benchmark rankings & pricing

Real-world SWE-bench scores & $20/mo vs API cost break-even calculator

🏆 Leaderboard→💰 Compare Plans↗
🔗

Project Links & Resources

Direct Access
Direct access to official project resources and documentation
🐙
gh-aw v0.90.1 release
github.com/github/gh-aw/releases/tag/v0.90.1
↗
🐙
Compare v0.90.0...v0.90.1
github.com/github/gh-aw/compare/v0.90.0...v0.90.1
↗
🐙
gh-aw README
github.com/github/gh-aw/blob/main/README.md
↗
📝
GitHub Next Agentic Workflows
githubnext.com/projects/agentic-workflows
↗
🐙
`v0.90.0`
github.com/github/gh-aw/releases/tag/v0.90.0
↗
🔬

In-Depth Technical Analysis

Background Agentic workflows that mutate issues/PRs need replayable ledgers and threat-detection artifacts. After v0.90.0’s friction-cost/noop graders, 0.90.1 hardens ledger independence and self-hosted runner policy. ### What shipped Standalone safe-output ledgers + replay (defaults kept on import); richer audit (txs, threat outcomes, evals, friction attribution); narrowed add-labels + required-labels on assign-to-agent; compile-time self-hosted runner enforcement; kebab-case sandbox frontmatter; Claude Code CLI 2.1.280. ### Benchmarks No SWE-bench. Anchor: 2026-09-30T17:20:00Z; compare v0.90.0...v0.90.1. Release note mentions ~13.3 AIC workflow self-metrics—not a cross-product bench. ### Get started Upgrade to v0.90.1; see https://github.com/github/gh-aw/blob/main/README.md and https://githubnext.com/projects/agentic-workflows