Build or update the BlueBubbles external channel plugin for Moltbot (extension package, REST...
npx skills add alistaircroll/verbose-deployment --skill "build"
Install specific skill from multi-skill repository
# Description
Run the production build including type-checking, and collect build metrics (duration, bundle size, route count, errors). Use as Phase 4 of the deployment pipeline or when verifying a project builds cleanly for production.
# SKILL.md
name: build
description: "Run the production build including type-checking, and collect build metrics (duration, bundle size, route count, errors). Use as Phase 4 of the deployment pipeline or when verifying a project builds cleanly for production."
Type Check & Build
Run the production build, which typically includes type-checking.
Detect
Look for build commands in package.json scripts or project config:
| Ecosystem | Commands to check |
|---|---|
| Node.js | npm run build, next build, vite build |
| TypeScript | tsc --noEmit (type check only) |
| Rust | cargo build --release |
| Go | go build ./... |
Why This Is Separate from Unit Tests
Many test runners (Vitest/esbuild, Jest/babel) skip type-checking for speed. The production build catches type errors that tests silently ignore. This is a real gap โ never skip this phase, even if all unit tests pass.
Example: a function parameter typed as string that receives number | undefined will pass Vitest (esbuild strips types) but fail next build or tsc --noEmit.
Run
Execute the detected build command. For TypeScript projects, if the build tool doesn't include type-checking, run tsc --noEmit separately first.
Collect
- Build duration
- Any type/compilation errors (file, line, message)
- Bundle/output size:
bash du -sh .next/ # Next.js du -sh dist/ # Vite du -sh target/release/ # Rust - Route count, page count, or equivalent metrics from build output
- Exit code
Stop Conditions
STOP if: Build fails. Fix type errors or compilation issues before proceeding.
# Supported AI Coding Agents
This skill is compatible with the SKILL.md standard and works with all major AI coding agents:
Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.