Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Manage and optimize monorepos with Turborepo, Nx, pnpm workspaces, and Changesets. Use when working in monorepos, running impact analysis, optimizing build times with remote caching, migrating from multi-repo, or coordinating publishing.
.claude/skills/borghei-monorepo-navigator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 197% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 66% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 164% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 84% | 0% |
Navigate, manage, and optimize monorepos at any scale. Covers Turborepo, Nx, pnpm workspaces, and Lerna/Changesets for cross-package impact analysis, selective builds on affected packages only, dependency graph visualization, remote caching configuration, migration from multi-repo to monorepo with preserved git history, and coordinated package publishing with automated changelogs.
Keywords: monorepo, Turborepo, Nx, pnpm workspaces, Changesets, dependency graph, remote cache, affected packages, selective builds, cross-package impact, npm publishing, workspace protocol
workspace:* replacement during publish.Before analyzing or configuring, confirm these inputs. If any is unknown or vague, ASK — do not assume:
dependency_graph.py vs impact_detector.py vs package_analyzer.py)--ref and what counts as changed)Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the artifact.
| Tool | Purpose | Command | |------|---------|---------| | dependency_graph.py | Generate a Mermaid dependency graph of internal packages | python scripts/dependency_graph.py . --focus @repo/ui --direction dependents | | impact_detector.py | Detect which packages are affected by file changes | python scripts/impact_detector.py . --ref origin/main --affected-only | | package_analyzer.py | Analyze package structure, dependencies, and health | python scripts/package_analyzer.py . --only-shared |
Load the reference that matches the task — keep this file lean and pull detail on demand:
This skill covers:
This skill does NOT cover:
performance-profilerci-cd-pipeline-builderrelease-managerdependency-auditor| Skill | Integration | Data Flow | |-------|------------|-----------| | ci-cd-pipeline-builder | Monorepo-aware CI workflows use --filter flags and remote caching tokens | Monorepo Navigator defines filter patterns and cache config that CI pipelines consume | | release-manager | Changesets versioning feeds into release orchestration and tag management | Release Manager triggers changeset version and changeset publish as part of release flow | | dependency-auditor | Workspace dependency graph informs vulnerability and license scanning scope | Monorepo Navigator exports the package dependency tree that Dependency Auditor analyzes | | performance-profiler | Build profiling data identifies slow packages for optimization | Performance Profiler measures per-package build times surfaced by Turborepo --summarize | | changelog-generator | Changesets produce per-package changelogs consumed by release notes | Changeset summaries flow into Changelog Generator for formatted release documentation | | tech-debt-tracker | Cross-package coupling and circular dependencies surface as tracked tech debt items | Monorepo Navigator's impact analysis identifies coupling hotspots that Tech Debt Tracker records |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 6,059 | 12,714 | +110% | 1 | 1 | 0% | 876 | 3,264 | +273% | 0 | 0 | — |
case-02 | fail→fail | 10,950 | 17,239 | +57% | 1 | 1 | 0% | 1,750 | 4,078 | +133% | 0 | 0 | — |
case-03 | fail→fail | 17,296 | 21,750 | +26% | 1 | 1 | 0% | 2,668 | 4,746 | +78% | 0 | 0 | — |
case-04 | fail→pass | 10,533 | 32,931 | +213% | 1 | 1 | 0% | 1,723 | 2,114 | +23% | 0 | 0 | — |
case-05 | fail→pass | 6,273 | 11,467 | +83% | 1 | 1 | 0% | 1,079 | 3,202 | +197% | 0 | 0 | — |
case-06 | fail→pass | 6,524 | 3,950 | -39% | 1 | 1 | 0% | 1,084 | 1,797 | +66% | 0 | 0 | — |
case-07 | fail→fail | 14,959 | 16,246 | +9% | 1 | 1 | 0% | 2,236 | 3,274 | +46% | 0 | 0 | — |
case-08 | pass→pass | 14,293 | 10,650 | -25% | 1 | 1 | 0% | 2,502 | 2,608 | +4% | 0 | 0 | — |
case-09 | pass→pass | 10,767 | 9,516 | -12% | 1 | 1 | 0% | 1,877 | 2,684 | +43% | 0 | 0 | — |
case-10 | pass→pass | 17,205 | 12,271 | -29% | 1 | 1 | 0% | 2,668 | 3,049 | +14% | 0 | 0 | — |
case-11 | pass→pass | 20,912 | 16,600 | -21% | 1 | 1 | 0% | 2,970 | 3,788 | +28% | 0 | 0 | — |
case-12 | pass→pass | 11,633 | 11,125 | -4% | 1 | 1 | 0% | 1,950 | 3,089 | +58% | 0 | 0 | — |
case-13 | pass→pass | 10,298 | 11,338 | +10% | 1 | 1 | 0% | 1,769 | 3,174 | +79% | 0 | 0 | — |
case-14 | pass→pass | 13,015 | 9,559 | -27% | 1 | 1 | 0% | 2,054 | 2,649 | +29% | 0 | 0 | — |
case-15 | pass→pass | 11,299 | 3,043 | -73% | 1 | 1 | 0% | 1,709 | 1,646 | -4% | 0 | 0 | — |
case-16 | pass→pass | 14,799 | 13,611 | -8% | 1 | 1 | 0% | 2,728 | 3,581 | +31% | 0 | 0 | — |
case-17 | pass→pass | 13,180 | 9,702 | -26% | 1 | 1 | 0% | 1,881 | 2,691 | +43% | 0 | 0 | — |
case-18 | fail→pass | 4,674 | 5,619 | +20% | 1 | 1 | 0% | 832 | 2,196 | +164% | 0 | 0 | — |
case-19 | fail→pass | 8,176 | 9,802 | +20% | 1 | 1 | 0% | 1,546 | 2,842 | +84% | 0 | 0 | — |
case-20 | fail→fail | 19,466 | 14,680 | -25% | 1 | 1 | 0% | 3,148 | 3,608 | +15% | 0 | 0 | — |
case-21 | fail→fail | 19,050 | 14,693 | -23% | 1 | 1 | 0% | 2,790 | 3,362 | +21% | 0 | 0 | — |
case-22 | fail→fail | 16,299 | 12,752 | -22% | 1 | 1 | 0% | 2,841 | 3,410 | +20% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +23 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.