Production hardening: ESLint, CI gate, coverage, honest LOC telemetry

[IgnazioDS]

Brings the dashboard to production grade. No product behavior changes — all gates now run and pass.

What changed

• ESLint configured (.eslintrc.json, next/core-web-vitals) — npm run lint now runs non-interactively and passes clean.
• Next.js CI gate (.github/workflows/nextjs-ci.yml): lint → type-check → coverage → build on every push/PR. Previously only Python had CI.
• Coverage tooling (@vitest/coverage-v8) with an enforced 80% threshold on src/lib/** (the logic layer). src/lib went from ~48% → 95% lines / 86% branches.
• New tests (51 total, was 39): api.ts client (happy/error/network), usePolling (load/error/reschedule/refetch/disabled), useHotkey, useAnimatedNumber easing loop, and project.ts metadata.
• Honest telemetry LOC: the static-LOC script was counting package-lock.json (~11.6k) and the generated coverage/ report (~5.7k), which would have inflated the public lines_of_code from a real ~5.8k to 24.8k. Excluded lockfiles + generated dirs; refreshed the artifact to 5778 (was a stale 1390 that predated the lockfile).
• Cleanup: dropped dead NexusRAG vars from .env.local.example (the showcase dashboard uses no env vars); added engines: node>=20; gitignored coverage/.

Verification (local)

• npm run lint ✓ · npm run type-check ✓ · npm run test:coverage ✓ (51 tests) · npm run build ✓
• Python: 9 tests ✓
• Telemetry endpoint shape unchanged; system: "runbook-orchestrator", mode: "showcase" still conformant to TELEMETRY_SCHEMA.md.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
agent-runbook-orchestrator	Ready	Preview, Comment	Jun 9, 2026 5:58pm

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.