speckit-resolve-prlisted

# Resolve PR Review Comments ## Capability discovery & grounding Before researching or recommending, enumerate the tools and skills your session actually exposes — do not assume a fixed set; the user may have installed anything — and select the best fit per `speckit-pro/skills/speckit-autopilot/references/capability-discovery.md`. Ground every external fact you assert in a real tool, skill, or file result per `speckit-pro/skills/speckit-autopilot/references/grounding.md`, and abstain when nothing grounds it. ## Input The user provides either: - A full PR URL: `https://github.com/owner/repo/pull/46` - A full PR review URL: `https://github.com/owner/repo/pull/46#pullrequestreview-123` - Just a PR number: `46` (repo detected from `git remote -v`) ## What to Do ### 1. Parse Input and Detect Repo ```text If full URL provided: Extract OWNER, REPO, PR_NUMBER from URL If just a number: Run `git remote -v` and extract OWNER/REPO from origin URL PR_NUMBER = the provided number ``` ### 2. Discover Project Commands Read CLAUDE.md and package.json (or equivalent) to find: - BUILD command - TYPECHECK command - LINT command - LINT_FIX command - UNIT_TEST command - INTEGRATION_TEST command Detect package manager from lockfile if Node.js project. ### 3. Fetch All Unresolved Review Threads Fetch review threads via GraphQL with `gh api graphql` to get thread IDs (needed for resolution) and comment details in one call. Query `repository.pullRequest.reviewThreads(first: 100)` a