feat: setup-context7 skill + using-context7 prerequisites pointer

Split context7 concerns into two skills:
- using-context7 (policy, every-time): when to call, how to query well
- setup-context7 (infrastructure, one-time): install plugin, inject API
  key, clean manual MCP entries, with confirmation gates

setup-context7 reuses an existing CONTEXT7_API_KEY by searching, in order:
~/.claude/settings.json → ~/.claude.json (top-level + project-scoped) →
~/.claude/settings.local.json → env. Asks the user only if nothing found.
Cross-platform (paths identical on Win/Linux/macOS; only JSON validator
differs). Phase-2 confirmation gate before any write; per-edit JSON
validation; rollback procedure documented.

using-context7 gets a small Prerequisites section pointing at
setup-context7 when MCP tools are missing.

Bonus fix: scripts/build.sh PS multi-arg bug. Comma-joined -Names didn't
split into [string[]] when build.ps1 was invoked via -File. Now the bash
side loops one PS invocation per skill.

Wiki: extended .wiki/concepts/context7-setup.md with the design rationale
(why split policy from setup) — template for future "X plugin + how to
use X" pairs. index.md and log.md updated.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

.tasks/STATUS.md

@@ -3,6 +3,16 @@ _Updated: 2026-04-28_
 
 ## Done
 
+### Create `setup-context7` skill
+- [x] Drafted `skills/setup-context7/SKILL.md` — 8 phases (discovery → plan → backup → install → inject → clean → smoke-test → restart) + rollback + cross-platform notes
+- [x] API-key discovery in priority order: `settings.json` → `.claude.json` (top-level + project-scoped) → `settings.local.json` → env → ask
+- [x] Phase 2 confirmation gate (no mutations without explicit "ok"); Phase 6 validates JSON after each edit
+- [x] Documented `/plugin update` gotcha
+- [x] Added Prerequisites section in `using-context7` pointing at `setup-context7`
+- [x] Built + installed both skills (`setup-context7` already in `available_skills`)
+- [x] Updated `.wiki/concepts/context7-setup.md` with "Now captured as a skill" section explaining the policy/setup split rationale
+- [x] **Bonus fix:** `scripts/build.sh` PS multi-arg bug — comma-joined `-Names` wasn't splitting into `[string[]]` when called via `-File`. Now loops in bash, one PS invocation per skill.
+
 ### Switch context7 from manual MCP config to official plugin
 - [x] Backup `~/.claude.json` and `~/.claude/settings.json` (suffix `.bak-20260428-114532`)
 - [x] User installed `context7@claude-plugins-official`