v0.6.0 — Governance Integration

Phase 4: Governance Integration (5/5)

• arbiter governance — repo maturity scoring: LICENSE, CONTRIBUTING, SECURITY, CoC, CI (#52)
• arbiter audit-trail — VERUM-aligned append-only hash-chained score records (#53)
• arbiter compliance — map findings to NIST CSF 2.0 + ISO 27001:2022 (#54)
• arbiter certify — full certification: CERTIFIED / PROVISIONAL / FAILED (#55)

Arbiter certifies itself

Certification: CERTIFIED
  Code Quality:  96.1 / 100
  Governance:    70.0 / 100
  Dependencies: 100.0 / 100
  Overall:       89.0 / 100

Stats

• 38+ CLI subcommands
• 600+ tests
• 33 PRs merged in one session
• Phases 1-4 complete — only Phase 5 (Platform) remains

🤖 Generated with Claude Code

v0.5.0 — Intelligence Layer

Phase 3: Intelligence Layer (6/6)

• arbiter suggest — rank findings by impact, recommend highest-ROI fixes (#48)
• arbiter explain — natural language score explanation (#49)
• arbiter profiles — enterprise/startup/oss/strict presets (#47)
• arbiter history — score tracking over time with deltas (#46)
• arbiter team — quality metrics by contributor (#51)
• arbiter deps — Python dependency scoring (#50)

Also in this release

• Preflight v2: try-import, notebook, comment-reexport checks — 100% false positive coverage (#44)
• Contribute v2: duplicate PR check + temp dir cleanup (#45)
• Learnings tracker: false positive and rejection ledger (#43)

Stats

• 35+ CLI subcommands
• 500+ tests
• 10 analyzers across 6 languages
• 29 PRs merged in one session
• 8 OSS contributions to top repos
• Phases 1-3 complete

🤖 Generated with Claude Code

v0.4.0 — Polyglot Contribution Engine

What's New

Phase 1: Contribution Engine

• arbiter contribute — automated scan → fix → prepare pipeline (#34)
• arbiter watch — GitHub API PR status tracking (#29)
• arbiter badge — SVG score badges for README.md (#32)
• .arbiter.yml / .arbiter.json — per-repo configuration (#31)
• CI quality gate — reusable GitHub Action workflow (#30)
• Benchmark dataset — 13 reference repos with score ranges (#33)

Phase 2: Polyglot Expansion

• Shell analyzer via shellcheck (#36)
• JavaScript/TypeScript analyzer via ESLint (#40)
• Go analyzer via golangci-lint (#40)
• Rust analyzer via cargo clippy (#40)
• Language detection for 25+ languages (#40)
• Per-language scoring weight profiles (#41)

Also in this release

• arbiter github-top — score popular GitHub repos by stars (#20)
• arbiter score-url — score any repo by GitHub URL (#21)
• arbiter compare — side-by-side repo comparison (#25)
• arbiter leaderboard — markdown quality index (#22)
• --noise-threshold — per-rule finding cap (#23)
• Weekly leaderboard GitHub Action (#24)
• CONTRIBUTING.md with DCO requirement (#26)
• Contribution tracker with JSONL ledger (#27)
• Product roadmap through 2027 (#28)

Stats

• 28 CLI subcommands
• 10 analyzers across 6 languages
• 427 tests
• 19 PRs merged in one session
• 6 OSS contributions submitted to top repos

---

Powered by HUMMBL — governed AI for enterprise.