Monorepo Navigator
Turborepo/Nx/pnpm workspace management and impact analysis.
$skillfish add ~skills/monorepo-navigator Real Conversations
See how the Monorepo Navigator skill responds to actual requests. These are representative examples of what your AI agent will produce.
Our Turborepo monorepo CI takes 22 minutes on every PR, even if I only changed one package. We have 6 apps and 12 packages. The GitHub Actions workflow runs turbo run build test lint with no filtering. How do I fix this?
You're rebuilding all 18 packages on every PR because you're not using Turborepo's affected detection. Here's the fix:
Problem: turbo run build test lint — no filter, rebuilds everything
Fix: filter to affected packages only
# .github/workflows/ci.yml
- uses: actions/checkout@v4
with:
fetch-depth: 0 # CRITICAL — without this, affected detection breaks
- name: Build affected
run: turbo run build --filter=...[origin/main]
- name: Test affected
run: turbo run test --filter=...[origin/main]
- name: Lint affected
run: turbo run lint --filter=...[origin/main]
...[origin/main] means: changed packages + everything that depends on them.
Add remote caching to make even full builds fast:
turbo login # authenticates to Vercel Remote Cache
turbo link # links this repo to a remote cache
Then in CI:
env:
TURBO_TOKEN: ${{ secrets.TURBO_TOKEN }}
TURBO_TEAM: ${{ vars.TURBO_TEAM }}
Expected result:
- PR touching 1 package: builds 1-3 packages (package + direct dependents)
- Cold build of everything: ~22min (unchanged)
- Warm build with remote cache hit: ~90 seconds for most PRs
Dry-run to see impact before merging:
turbo run build --filter=...[origin/main] --dry-run
# Shows: "Would run 3 tasks" vs current "18 tasks"
Get Started
How to install and use this skill in your preferred environment.
Skills are designed for AI coding agents (Claude Code, Cursor, Windsurf) and IDE-based workflows where the agent can read files, run scripts, and act on your codebase. Web-based AI can use the knowledge and frameworks, but won't have tool access.
Models & Context
Which AI models and context windows work best with this skill.
Recommended Models
Larger models produce more detailed, production-ready outputs.
Context Window
This skill's SKILL.md is typically 3–10 KB — fits in any modern context window.
All current frontier models (Claude, GPT, Gemini) support 100K+ context. Use the full window for complex multi-service work.
Pro tips for best results
Be specific
Include numbers — users, budget, RPS — so the skill can size the architecture.
Share constraints
Compliance needs, team size, and existing stack all improve the output.
Iterate
Start with a high-level design, then ask follow-ups for IaC, cost analysis, or security review.
Combine skills
Pair with companion skills below for end-to-end coverage.
Ready to try Monorepo Navigator?
Install the skill and start getting expert-level guidance in your workflow — any agent, any IDE.
$skillfish add ~skills/monorepo-navigator