migrate-hardhat2-to-hardhat3

name	migrate-hardhat2-to-hardhat3
description	Systematically migrate a Hardhat V2 project to Hardhat V3, covering core changes (ESM, declarative config, hooks system, network connections) and the full plugin ecosystem (hardhat-deploy, hardhat-verify, @openzeppelin/hardhat-upgrades, and others), coverage tools, and test migration. Use when upgrading any production Hardhat V2 repository to V3.

This skill runs the full V2→V3 migration as six sequential phases. A single agent running them top-to-bottom is the default — it handles small and medium projects (and large ones up to ~150 .ts test files on a 1M-token context window). Beyond that, or on a smaller-context harness (~200k tokens) past a trivial project, fan the file-heavy phases (4 and 5) out to sub-agents so a single context doesn't overflow mid-migration (see "Optional: sub-agent fan-out" in references/migration-conventions.md).

Before you start

1. Read references/migration-conventions.md — the rules every phase depends on (no VCS commands, package-manager detection, scoping, gates, change presentation, and the optional parallel-execution model). Don't proceed until you have.
2. Detect the package manager (lockfile check) and remember it for the whole migration.
3. Choose the execution mode. Estimate your context window and the project size (count .ts test files) and pick single-agent (the default) vs. sub-agent fan-out using the thresholds in the conventions doc. If you can't tell which window you're on, assume the smaller one. Most projects stay single-agent.
4. Phases are strictly sequential — each depends on the previous one's output. After every phase, stop at a user checkpoint and wait for approval before continuing. Present the checkpoint in this order:
   • Step done — the title of the phase just completed (e.g. ) and its changes: list the changed files and surface any unexpected ones verbatim (see core rule 5 in the conventions doc).