refactor(rules): Split skill-agent-updates into rule + reference doc

- Trim rule to minimal "check docs first" requirement (10 lines)
- Move detailed reference to docs/SKILL-SUBAGENT-REFERENCE.md

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

docs/SKILL-SUBAGENT-REFERENCE.md

@@ -0,0 +1,55 @@
+# Skill and Subagent Reference
+
+Quick reference for Claude Code skill and subagent APIs. **Always check official docs first** - this may be outdated.
+
+## Skill Frontmatter Fields (January 2026)
+
+```yaml
+---
+name: skill-name                    # Required: kebab-case
+description: "Triggers on: ..."     # Required: include trigger keywords
+allowed-tools: "Read Write Bash"    # Restrict available tools
+disable-model-invocation: false     # true = manual /skill only
+user-invocable: true                # false = hide from slash completion
+context: main                       # main | fork (subagent isolation)
+agent: custom-agent                 # Custom system prompt agent
+hooks:
+  preToolUse:
+    - command: "echo pre"
+  postToolUse:
+    - command: "echo post"
+---
+```
+
+## Subagent Options
+
+| Field | Values | Purpose |
+|-------|--------|---------|
+| `permissionMode` | default, acceptEdits, bypassPermissions | Control autonomy |
+| `skills` | [skill-names] | Preload skills in subagent |
+| `model` | sonnet, opus, haiku | Override model |
+
+## Decision Framework: Main Context vs Fork
+
+| Question | If Yes → | If No → |
+|----------|----------|---------|
+| Needs current session state (tasks, conversation)? | Main context | Consider fork |
+| Output verbose (>500 lines)? | Consider fork | Main context |
+| Needs user interaction during execution? | Main context | Consider fork |
+| One-shot research/analysis task? | Fork | Main context |
+
+## Skills Using Subagent Isolation
+
+| Skill | Method | Why |
+|-------|--------|-----|
+| `introspect` | Task agent (background) | Session log analysis is verbose |
+
+## Session Commands Analysis
+
+| Command | Context | Rationale |
+|---------|---------|-----------|
+| `/sync` | Main | Must restore session state (tasks, context) |
+| `/save` | Main | Must access current tasks via TaskList |
+| `/canvas` | Main | Interactive TUI requires real-time feedback |
+
+These MUST run in main context - subagent isolation would break their core functionality.

rules/skill-agent-updates.md

@@ -1,66 +1,10 @@
 # Skill and Agent Updates
 
-## Mandatory Documentation Check
+**BEFORE creating or updating any skill or agent**, check the official docs:
 
-**BEFORE creating or updating any skill or agent**, always check the official Claude Code documentation:
+| Resource | URL |
+|----------|-----|
+| Skills | https://code.claude.com/docs/en/skills |
+| Sub-agents | https://code.claude.com/docs/en/sub-agents |
 
-| Resource | URL | Check For |
-|----------|-----|-----------|
-| Skills | https://code.claude.com/docs/en/skills | New frontmatter fields, context options |
-| Sub-agents | https://code.claude.com/docs/en/sub-agents | Permission modes, built-in agents |
-
-These docs change frequently. Features we should watch for:
-
-## Current Skill Frontmatter Fields (January 2026)
-
-```yaml
----
-name: skill-name                    # Required: kebab-case
-description: "Triggers on: ..."     # Required: include trigger keywords
-allowed-tools: "Read Write Bash"    # Restrict available tools
-disable-model-invocation: false     # true = manual /skill only
-user-invocable: true                # false = hide from slash completion
-context: main                       # main | fork (subagent isolation)
-agent: custom-agent                 # Custom system prompt agent
-hooks:
-  preToolUse:
-    - command: "echo pre"
-  postToolUse:
-    - command: "echo post"
----
-```
-
-## Current Subagent Options
-
-| Field | Values | Purpose |
-|-------|--------|---------|
-| `permissionMode` | default, acceptEdits, bypassPermissions | Control autonomy |
-| `skills` | [skill-names] | Preload skills in subagent |
-| `model` | sonnet, opus, haiku | Override model |
-
-## Decision Framework: Main Context vs Fork
-
-| Question | If Yes → | If No → |
-|----------|----------|---------|
-| Does it need current session state (tasks, conversation)? | Main context | Consider fork |
-| Is output verbose (>500 lines)? | Consider fork | Main context |
-| Does it need user interaction during execution? | Main context | Consider fork |
-| Is it a one-shot research/analysis task? | Fork | Main context |
-
-## Skills Using Subagent Isolation
-
-Skills that delegate to Task agents or use `context: fork`:
-
-| Skill | Method | Why |
-|-------|--------|-----|
-| `introspect` | Task agent (background) | Session log analysis is verbose |
-
-## Session Commands Analysis
-
-| Command | Context | Rationale |
-|---------|---------|-----------|
-| `/sync` | Main | Must restore session state (tasks, context) |
-| `/save` | Main | Must access current tasks via TaskList |
-| `/canvas` | Main | Interactive TUI requires real-time feedback |
-
-These MUST run in main context - subagent isolation would break their core functionality.
+These APIs change frequently. For detailed reference (frontmatter fields, decision frameworks), see `docs/SKILL-SUBAGENT-REFERENCE.md`.