📦 Add semantic versioning to all files (#11)

* 📦 Add semantic versioning to all 55 files

Implement per-file semantic versioning (version: X.Y.Z) in frontmatter for skills,
commands, agents, rules, and personalities. Versions determined from Git history
analysis:

- 28 files @ 1.0.0 (production-ready, 2+ months stable)
- 7 files @ 0.x (pre-release, actively evolving)
- 3 files @ 1.2.0 (marketplace-distributed)
- 1 file @ 0.1.0 (brand new, < 2 weeks old)
- 2 files @ 2.x (foundational rules with major versions)

Update ai-coding-config.md command with version comparison logic for selective
updates. Users will be offered updates only when source version > installed version.

Consistent versioning criteria:
- 0.1.0: Genuinely new (< 2 weeks)
- 0.2-0.4: Pre-release with refinements
- 1.0.0+: Production-ready or 2+ months with stable history
- 2.0+: Major architectural/consolidation changes

🧠 Generated with Claude Code

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* 🔧 Fix descriptions to single-line format for Claude parsing

Move all multi-line descriptions to single-line frontmatter format. Claude requires
descriptions on same line as "description:" key - prettier cannot reformat to
multiple lines.

Descriptions condensed to stay under 88 chars to prevent prettier wrapping:
- Commands: generate-AGENTS-file, handoff-context, product-intel, create-prompt, autotask, generate-llms-txt
- Personalities: marie-kondo, samantha, sherlock, stewie
- Rules: heart-centered-ai-philosophy, trust-and-decision-making, user-facing-language

🧠 Generated with Claude Code

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Nick Sullivan <nick@technick.ai>
Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

Author: 3 people

plugins/agents/agents/debugger.md

@@ -1,6 +1,7 @@
 ---
 name: debugger
 description: "Invoke for errors or unexpected behavior"
+version: 1.0.0
 ---
 
 I hunt root causes, not symptoms. I don't slap band-aids on problems - I find out why

plugins/agents/agents/design-reviewer.md

@@ -2,6 +2,7 @@
 name: design-reviewer
 description: "Invoke after frontend code changes for design review"
 color: pink
+version: 1.0.0
 ---
 
 <identity>

plugins/agents/agents/git-writer.md

@@ -2,6 +2,7 @@
 name: git-writer
 description: "Use proactively for commits, PRs, and branch names"
 model: haiku
+version: 0.2.0
 ---
 
 I'm Git Writer, and I write git messages that make future developers thank you 📚. I

plugins/agents/agents/prompt-engineer.md

@@ -2,6 +2,7 @@
 name: prompt-engineer
 description: "Invoke when creating agent prompts or LLM instructions"
 model: opus
+version: 0.3.0
 ---
 
 I speak fluent LLM. I craft prompts that work WITH how language models actually process

plugins/agents/agents/seo-specialist.md

@@ -1,6 +1,7 @@
 ---
 name: seo-specialist
 description: "Invoke for SEO audits and optimization"
+version: 1.0.0
 ---
 
 <identity>

plugins/agents/agents/site-keeper.md

@@ -1,6 +1,7 @@
 ---
 name: site-keeper
 description: "Invoke for production health monitoring and error triage"
+version: 1.0.0
 ---
 
 <identity>

plugins/agents/agents/test-runner.md

@@ -2,6 +2,7 @@
 name: test-runner
 description: "Invoke to run tests with terse, context-efficient results"
 model: haiku
+version: 0.1.0
 ---
 
 I run tests and tell you exactly what you need to know. Pass count. Fail count. For

plugins/agents/agents/ux-designer.md

@@ -1,6 +1,7 @@
 ---
 name: ux-designer
 description: "Invoke for user-facing content and interface design"
+version: 0.4.0
 ---
 
 I notice everything. Every unnecessary word. Every weak verb. Every moment of friction.

plugins/core/commands/address-pr-comments.md

@@ -2,6 +2,7 @@
 description: Triage and address PR comments from code review bots intelligently
 argument-hint: [pr-number]
 model: sonnet
+version: 1.0.0
 ---
 
 # Address PR Comments

plugins/core/commands/ai-coding-config.md

@@ -1,6 +1,7 @@
 ---
 description: Set up, update, or add AI coding configurations
 argument-hint: [update | add]
+version: 1.0.0
 ---
 
 # AI Coding Configuration
@@ -305,32 +306,42 @@ If direct symlinks to deleted paths found, offer to update:
 </symlink-compatibility-check>
 
 <file-updates>
-For Cursor/Windsurf, update by category:
+For Cursor/Windsurf, update COPIED files (rules, personalities) using version comparison. Symlinked files (commands, agents, skills) are already current from git pull.
 
-1. **Rules** (`rules/`)
-   - Compare each rule file in `~/.ai_coding_config/rules/` vs project
-   - Show diffs for significant changes
-   - Preserve project-specific rules
+Each file has a `version: X.Y.Z` field in frontmatter. Missing version = v0.0.0. Offer updates only when source version is newer than installed version.
 
-2. **Commands/Agents/Skills**
-   - If symlinked to `~/.ai_coding_config/plugins/`, already updated via git pull
-   - If copied, compare and offer updates
+When updates are available, use AskUserQuestion:
 
-3. **Personalities** (`rules/personalities/`)
-   - Check if personality has updates
-   - Preserve `alwaysApply` setting
+"Updates available:
+- git-interaction.mdc: 1.0.0 → 1.1.0
+- prompt-engineering.mdc: 1.0.0 → 1.2.0
+- new-rule.mdc (new, v1.0.0)
 
-Present update strategy: "Update all", "Update selectively", or custom.
-Never silently overwrite project customizations.
+12 files already current."
+
+Options:
+- Update all
+- Select individually
+- Show diffs first
+- Skip updates
+
+When everything is current: "All files are up to date. No updates needed."
+
+For personalities, preserve the user's `alwaysApply` setting when updating content.
 </file-updates>
 
 </cursor-windsurf-update>
 
 <update-summary>
-Report what was updated:
-- "Updated X plugins to version Y"
-- "Updated Z rule files"
-- "No updates available" (if already current)
+For Claude Code:
+"Updated core, agents, skills plugins to version 1.2.0"
+
+For Cursor/Windsurf:
+"Update complete:
+- git-interaction.mdc: 1.0.0 → 1.1.0
+- prompt-engineering.mdc: 1.0.0 → 1.2.0
+- Installed new-rule.mdc (v1.0.0)
+- 12 files already current"
 </update-summary>
 
 </update-mode>
@@ -360,38 +371,41 @@ Clarifying questions:
 </mechanism-selection>
 
 <artifact-creation>
-All new artifacts go in the appropriate plugin directory:
+All new artifacts go in the appropriate plugin directory with `version: 1.0.0`:
 
-**Commands**: Create in `plugins/core/commands/command-name.md`
+**Commands**: `plugins/core/commands/command-name.md`
 ```yaml
 ---
 description: Brief explanation
 argument-hint: [optional args]
+version: 1.0.0
 ---
 ```
 
-**Skills**: Create `plugins/skills/skills/skill-name/SKILL.md`
+**Skills**: `plugins/skills/skills/skill-name/SKILL.md`
 ```yaml
 ---
 name: skill-name
 description: "Use when [trigger]. Does [what] to achieve [outcome]."
+version: 1.0.0
 ---
 ```
 
-**Agents**: Create in `plugins/agents/agents/agent-name.md`
+**Agents**: `plugins/agents/agents/agent-name.md`
 ```yaml
 ---
 name: agent-name
 description: "Invoke when [trigger]"
 tools: Read, Write, Edit, Grep, Glob, Bash
 model: sonnet
+version: 1.0.0
 ---
 ```
 
-**Personalities**: Create `plugins/personalities/personality-name/`
-- `personality.mdc` - The personality definition
-- `.claude-plugin/plugin.json` - Plugin manifest
-- Update marketplace.json to include new personality
+**Personalities**: `plugins/personalities/personality-name/`
+- `personality.mdc` with `version: 1.0.0` in frontmatter
+- `.claude-plugin/plugin.json`
+- Update marketplace.json
 </artifact-creation>
 
 <creation-verification>