ln-821-npm-upgrader
Upgrades npm/yarn/pnpm dependencies with breaking change handling. Use when updating JavaScript/TypeScript dependencies.
- License 许可
- MIT
- 最近更新
- 2026-06-15
- Stars
- ★ 2
- Forks
- 1
01 前言 Front Matter
- name
- ln-821-npm-upgrader
- license
- MIT
- description
- Upgrades npm/yarn/pnpm dependencies with breaking change handling. Use when updating JavaScript/TypeScript dependencies.
02 资源文件 3
- 参考文档references/breaking_changes_patterns.md3.9 KB
- 参考文档references/coordinator_summary_contract.md1009 B
- 参考文档references/npm_peer_resolution.md2.0 KB
03 SKILL.md 正文
Paths: File paths (
references/,../ln-*) are relative to this skill directory.
ln-821-npm-upgrader
Type: L3 Worker Category: 8XX Optimization
Upgrades Node.js dependencies using npm, yarn, or pnpm with automatic breaking change detection and migration.
Overview
| Aspect | Details |
|---|---|
| Input | Project path, package manager type, upgrade policy |
| Output | Updated package manifest and a machine-readable dependency upgrade summary |
| Supports | npm, yarn (classic and berry), pnpm |
Workflow
Phases: Pre-flight -> Analyze -> Security Audit -> Check Outdated -> Identify Breaking -> Apply Upgrades -> Apply Migrations -> Verify Build -> Report
Phase 0: Pre-flight Checks
| Check | Required | Action if Missing |
|---|---|---|
package.json |
Yes | Block upgrade |
Lock file (package-lock.json, yarn.lock, pnpm-lock.yaml) |
No | Warn and regenerate before final verification |
| Package manager available | Yes | Block upgrade |
| Workspace baseline safe | Yes | In managed runs coordinator already prepared it; in standalone runs protect rollback locally |
Runtime Coordination
Managed runs receive deterministic runId and exact summaryArtifactPath from ln-820.
Standalone runs remain supported; if runtime arguments are omitted, generate a standalone run-scoped artifact before returning.
Phase 1: Analyze Dependencies
Read package.json and categorize dependencies for upgrade priority.
| Category | Examples | Priority |
|---|---|---|
| peer | typescript, @types/* |
1 |
| framework | react, vue, next |
2 |
| build | vite, webpack, esbuild |
3 |
| ui | @radix-ui/*, tailwindcss |
4 |
| state | @tanstack/react-query, zustand |
5 |
| utils | lodash, date-fns |
6 |
| dev | eslint, prettier, test tooling |
7 |
Phase 2: Security Audit
| Manager | Command |
|---|---|
| npm | npm audit --audit-level=high |
| yarn | yarn audit --level high |
| pnpm | pnpm audit --audit-level high |
Actions:
| Severity | Action |
|---|---|
| Critical | Block and report |
| High | Warn and continue |
| Moderate/Low | Log only |
Phase 3: Check Outdated
| Manager | Command |
|---|---|
| npm | npm outdated --json |
| yarn | yarn outdated --json |
| pnpm | pnpm outdated --json |
Phase 4: Identify Breaking Changes
MANDATORY READ: Load breaking_changes_patterns.md for full patterns.
Detection flow:
- Compare current vs latest major versions.
- Check shared breaking-change patterns.
- Query Context7 or Ref for migration guides before changing code.
Common breaking examples:
| Package | Breaking Version | Key Changes |
|---|---|---|
| react | 18 -> 19 | JSX transform, refs as props |
| vite | 5 -> 6 | ESM-only, newer Node baseline |
| eslint | 8 -> 9 | Flat config |
| tailwindcss | 3 -> 4 | CSS-first config |
| typescript | 5.4 -> 5.5+ | Stricter inference |
Phase 5: Apply Upgrades
Upgrade order:
- peer dependencies
- framework packages
- build tools
- UI libraries
- utilities
- dev dependencies
| Manager | Command |
|---|---|
| npm | npm install <package>@latest --save |
| yarn | yarn add <package>@latest |
| pnpm | pnpm add <package>@latest |
Peer dependency conflicts:
| Situation | Solution |
|---|---|
| ERESOLVE | npm install --legacy-peer-deps |
| Still fails | npm install --force only as last resort |
MCP Tools for Migration Search
| Priority | Tool | When to Use |
|---|---|---|
| 1 | mcp__context7__query-docs |
First choice for library docs |
| 2 | mcp__Ref__ref_search_documentation |
Official docs and GitHub |
| 3 | WebSearch | Latest info and community fixes |
Use MCP tools to fetch migration guides before applying non-trivial changes.
Phase 6: Apply Migrations
- Use MCP tools to find the current migration guide.
- Apply automated code transforms only when the guide supports them.
- Log manual follow-up steps for the final report.
Do not hardcode migrations without checking current documentation.
Phase 7: Verify Build
| Check | Command |
|---|---|
| TypeScript | npm run check or npx tsc --noEmit |
| Build | npm run build |
| Tests | npm test if available |
On failure:
- Identify the failing package.
- Search Context7 or Ref for the fix.
- If unresolved, rollback that package and continue with the remaining candidates.
Phase 8: Report Results
| Field | Description |
|---|---|
project |
Project path |
packageManager |
npm, yarn, or pnpm |
duration |
Total time |
upgrades.major[] |
Breaking changes applied |
upgrades.minor[] |
Feature updates |
upgrades.patch[] |
Bug fixes |
migrations[] |
Applied migrations |
skipped[] |
Already latest or policy-skipped |
verification |
Build/test/type-check verdict |
warnings[] |
Non-blocking issues |
artifact_path |
Durable worker report path, if written |
Configuration
Options:
upgradeType: major # major | minor | patch
allowBreaking: true
autoMigrate: true
queryMigrationGuides: true
auditLevel: high
minimumReleaseAge: 14
legacyPeerDeps: false
force: false
runBuild: true
runTests: false
runTypeCheck: true
rollbackOnFailure: true
Error Handling
| Error | Cause | Solution |
|---|---|---|
| ERESOLVE | Peer dependency conflict | Retry with legacy peer dependency mode |
| ENOENT | Missing lock file | Regenerate dependencies first |
| Build fail | Breaking change | Apply migration guide or rollback offending package |
| Type errors | Version mismatch | Update types or framework peer packages |
Rollback:
Restore package.json and the lock file, then run a clean install to restore the previous state.
References
Runtime Summary Artifact
MANDATORY READ: Load references/coordinator_summary_contract.md
Emit a dependency-worker summary envelope.
Managed mode:
ln-820passes deterministicrunIdand exactsummaryArtifactPath- write the summary to the provided
summaryArtifactPath
Standalone mode:
- omit
runIdandsummaryArtifactPath - write
.hex-skills/runtime-artifacts/runs/{run_id}/dependency-worker/ln-821--{identifier}.json
Monitor (2.1.98+): For install/audit/build/test commands expected >30s, use Monitor. Fallback: Bash(run_in_background=true).
Definition of Done
- [ ] Package manifest analyzed and dependencies prioritized
- [ ] Security audit completed for the selected package manager
- [ ] Outdated packages identified
- [ ] Breaking changes checked via patterns plus current docs
- [ ] Upgrades applied with rollback on failure
- [ ] Build and relevant verification commands pass after upgrades
- [ ]
dependency-workersummary artifact written to the managed or standalone path
Version: 1.1.0 Last Updated: 2026-01-10
04 同仓库更多 Skills
solana-dev
Use when user asks to "build a Solana dapp", "write an Anchor program", "create a token", "debug Solana errors", "set up wallet connection", "test my Solana program", "deploy to devnet", or "explain Solana concepts" (rent, accounts, PDAs, CPIs, etc.). End-to-end Solana development playbook covering wallet connection, Anchor/Pinocchio programs, Codama client generation, LiteSVM/Mollusk/Surfpool testing, and security checklists. Integrates with the Solana MCP server for live documentation search. Prefers framework-kit (@solana/client + @solana/react-hooks) for UI, wallet-standard-first connection (incl. ConnectorKit), @solana/kit for client/RPC code, and @solana/web3-compat for legacy boundaries.
generating-trading-signals
Generate trading signals using technical indicators (RSI, MACD, Bollinger Bands, etc.). Combines multiple indicators into composite signals with confidence scores. Use when analyzing assets for trading opportunities or checking technical indicators. Trigger with phrases like "get trading signals", "check indicators", "analyze for entry", "scan for opportunities", "generate buy/sell signals", or "technical analysis".
prediction-monitor
Prediction Monitor - Auto-activating skill for ML Deployment. Triggers on: prediction monitor, prediction monitor Part of the ML Deployment skill category.
strategy-advisor
High-level strategic thinking and business decision guidance for planning and direction-setting. Use when: making strategic decisions, evaluating business options, setting direction, analyzing trade-offs, or when user mentions strategy, business planning, competitive analysis, or long-term planning.
app-paywall-pilot
Design, audit, debug, and implement App Store-compliant mobile paywalls, subscription flows, pricing screens, trials, offers, feature gates, and subscription lifecycle UX. Use when the user asks to audit my paywall, fix App Store 3.1.2 rejection, improve trial-to-paid, choose plans/prices/trials, review RevenueCat/Adapty/Apphud/Superwall setup, calculate LTV/ROAS, or build paywall implementation.
figma
Use the Figma MCP server to fetch design context, screenshots, variables, and assets from Figma, and to translate Figma nodes into production code. Trigger when a task involves Figma URLs, node IDs, design-to-code implementation, or Figma MCP setup and troubleshooting.