fix(rules,plugin): normalize skill ids and fix shortcut recommendations (#1243, #1247)

[JeremyDev87]

Summary

• Remove superpowers: prefix from skill references in solution-architect.json and technical-planner.json agent definitions
• Replace _SHORTCUT_MAP in prompt_pattern_analyzer.py with real commands (/plan, /act, /eval) and bare skill names — removes all superpowers:* and oh-my-claudecode:* namespace references

Closes #1243
Closes #1247

Test plan

• grep -r "superpowers:" packages/rules/.ai-rules/agents/{solution-architect,technical-planner}.json — no matches
• grep -r "superpowers:\|oh-my-claudecode:" packages/claude-code-plugin/hooks/lib/prompt_pattern_analyzer.py — no matches
• python3 -m pytest hooks/tests/test_prompt_pattern_analyzer.py -v — 44 passed

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
codingbuddy-landing	Ready	Preview, Comment	Apr 3, 2026 6:16pm

[JeremyDev87]

Review: ✅ APPROVE

CI Status: ALL PASS

Code Review

prompt_pattern_analyzer.py — All 11 _SHORTCUT_MAP entries fixed:

• Shortcuts now map to real commands: /plan, /act, /eval (all exist in commands/)
• Skill names are bare catalog names: test-driven-development, systematic-debugging, brainstorming, etc.
• No more superpowers: or oh-my-claudecode: prefixes

solution-architect.json — 7 superpowers: references removed:

• Skill names, mandatory_checklist, verification_guide, execution_order, reference section all updated

technical-planner.json — 11 superpowers: references removed:

• Same thorough treatment across skills, mandatory_checklist, verification_guide, execution_order, execution_options, reference

Issues Found: None

Recommendation: APPROVE