oh-my-customcode 1.1.0 → 1.1.2

package/templates/.claude/skills/post-release-followup/SKILL.md

@@ -27,15 +27,9 @@ Gather unfinished work from multiple sources:
 
 ### Tool: Writing artifacts under .claude/outputs/
 
-CC sensitive-path check inspects tool target paths and triggers permission prompts on `.claude/` regardless of `bypassPermissions` and allow rules (refs: #960, #961, #978, #981, #1016).
+Under `mode: "bypassPermissions"`, subagents write directly to `.claude/outputs/sessions/` with the Write tool — direct `.claude/**` writes are permitted (CC v2.1.121+, #1101). No `/tmp` staging or script wrapping is needed. Read-only Bash on `.claude/outputs/` (e.g., `cat`, `head`, `wc`) is allowed for verification.
 
-To write post-release-followup results under `.claude/outputs/sessions/`:
-
-1. Write the artifact body to `/tmp/post-release-followup-$(date +%H%M%S).md` first (Write tool target = `/tmp`, no sensitive-path trigger)
-2. Use a `/tmp/*.sh` Bash script to move/copy the file under `.claude/outputs/sessions/$(date +%Y-%m-%d)/` (Bash target = `/tmp`, script-internal `cp` to `.claude/` is not audited)
-3. Read-only Bash on `.claude/outputs/` (e.g., `cat`, `head`, `wc`) is allowed for verification
-
-Reference: `feedback_sensitive_path_tmp_bypass.md`, R006 sensitive-path handling, #1016, #1045.
+Reference: R006/R010 sensitive-path handling (direct `.claude/**` write under bypassPermissions), #1101.
 
 - Extract any MEDIUM or LOW severity findings that were flagged but not fixed
 

package/templates/.claude/skills/professor-triage/SKILL.md

@@ -50,7 +50,7 @@ Analyzes GitHub issues directly against the current codebase. For each issue, se
 | Phase 4E artifact report | general-purpose | bypassPermissions |
 | Phase 5 GitHub actions | mgr-gitnerd | bypassPermissions |
 
-**Agent selection constraint**: Phases 4A, 4B, 4C, 4E MUST use `general-purpose` (NOT `arch-documenter`). `arch-documenter` has `disallowedTools: [Bash]` — cannot execute `/tmp/*.sh` bypass → falls back to Write → triggers CC sensitive-path guard on `.claude/outputs/`. See #1043.
+**Agent selection constraint**: Phases 4A, 4B, 4C, 4E MUST use `general-purpose` (NOT `arch-documenter`). `arch-documenter` has `disallowedTools: [Bash]` — cannot run the `gh`/shell commands these phases require. See #1043.
 
 ## Parallelization (R009/R018)
 
@@ -59,24 +59,20 @@ Analyzes GitHub issues directly against the current codebase. For each issue, se
 - 10+ issues: Agent Teams per R018
 - Phase 4A + 4B: parallel; Phase 4C: after both; Phase 4D + 4E: parallel; Phase 4F: gate
 
-## Sensitive-Path Artifact Protocol (MANDATORY)
+## Artifact Output (R006/R010)
 
-**R010 Universal /tmp Script Bypass (#1052)**: ALL `.claude/` modifications MUST use `/tmp/*.sh` script via Bash. Direct Write/Edit/Bash on `.claude/` triggers CC sensitive-path guard regardless of bypassPermissions.
-
-When spawning Phase 4A/4B/4C/4E agents, include verbatim in each agent prompt:
+Under `mode: "bypassPermissions"`, agents write directly to `.claude/outputs/` with Write/Edit — no temp-script wrapping is needed (CC v2.1.121+, #1101). When spawning Phase 4A/4B/4C/4E agents, include verbatim in each agent prompt:
 
 ```
-**Sensitive-path artifact protocol (mandatory for this task)**
-
-If your task involves writing artifacts under `.claude/outputs/`:
-1. Build artifact body in /tmp first: `cat > /tmp/professor-triage-<timestamp>.sh <<'EOF' ... EOF`
-2. Script content: `mkdir -p .claude/outputs/sessions/<date>/ && cat > .claude/outputs/sessions/<date>/<artifact>.md <<'ARTIFACT' ... ARTIFACT`
-3. Execute: `bash /tmp/professor-triage-<timestamp>.sh`
-4. Cleanup: `rm /tmp/professor-triage-<timestamp>.sh`
-DO NOT use Write/Edit directly on `.claude/outputs/` — CC sensitive-path guard triggers regardless of bypassPermissions/allow rules.
+**Artifact output (for this task)**
+
+If your task involves writing artifacts under `.claude/outputs/`, write the file to
+`.claude/outputs/sessions/<date>/<artifact>.md` using the Write tool directly (create the
+directory if needed). Your Agent tool call runs under `mode: "bypassPermissions"`, so direct
+Write/Edit on `.claude/` is permitted.
 ```
 
-See R006 "Sensitive Path Handling" + `feedback_sensitive_path_tmp_bypass.md`.
+See R006 "Sensitive Path Handling" (CC v2.1.121+ direct-write convention).
 
 ## Phase 5 Action Policy
 

package/templates/.claude/skills/qa-lead-routing/SKILL.md

@@ -5,11 +5,11 @@ scope: core
 user-invocable: false
 context: fork
 ---
-## Mandatory delegation directive (R010 Universal /tmp Script Bypass)
+## Mandatory delegation directive (R010 bypassPermissions)
 
 When this skill spawns a subagent via the Agent tool, the spawned prompt MUST include this directive verbatim (or equivalent):
 
-> ALL tool calls targeting `.claude/` paths MUST go through `/tmp/*.sh` script via Bash. This includes read-only measurement (sed/cat/wc/ls/grep), Write, Edit. Read tool is exempt. Direct Write/Edit/Bash on .claude/ triggers user approval prompts that block unattended automation. See R010 for the full pattern.
+> 이 스킬이 Agent tool로 서브에이전트를 스폰할 때 mode: "bypassPermissions"를 반드시 포함한다. 서브에이전트는 .claude/ 경로에 직접 Write/Edit/Bash를 사용한다(/tmp 우회 불필요, CC v2.1.121+). See R010.
 
 This directive is preserved inline because Agent-tool prompt synthesis can drop SKILL.md notes; inline mandatory directives survive (#1046 lesson).
 

package/templates/.claude/skills/reasoning-sandwich/SKILL.md

@@ -35,7 +35,7 @@ A model allocation pattern that wraps implementation actions with stronger-model
 | Phase | Recommended Model | Rationale |
 |-------|------------------|-----------|
 | Pre-reasoning (analyze/plan) | opus | Complex architectural reasoning, edge case detection |
-| Pre-reasoning (Opus 4.7) | opus47 | Opus 4.7 makes fewer tool calls and reasons more deeply; pre-reasoning phase should include explicit tool batch planning |
+| Pre-reasoning (Opus 4.8) | opus48 | Opus 4.8 makes fewer tool calls and reasons more deeply; pre-reasoning phase should include explicit tool batch planning |
 | Action (implement/generate) | sonnet | Optimized for code generation, balanced cost |
 | Post-verification (review/test) | sonnet or haiku | Structural verification, checklist validation |
 
@@ -63,23 +63,23 @@ This pattern is used by:
 | Opus for everything | Wasteful, slow | Reserve opus for reasoning-heavy phases |
 | Haiku for planning | Insufficient depth | Use opus for complex analysis |
 | Skipping verification | False completion risk | Always include post-verification phase |
-| Forcing frequent tool calls on Opus 4.7 | Fights model's natural reasoning-first tendency | Let Opus 4.7 reason; batch tool calls in pre-reasoning |
+| Forcing frequent tool calls on Opus 4.8 | Fights model's natural reasoning-first tendency | Let Opus 4.8 reason; batch tool calls in pre-reasoning |
 
-## Opus 4.7 Considerations
+## Opus 4.8 Considerations
 
-Opus 4.7 exhibits a distinct behavioral pattern: fewer tool calls with deeper reasoning per call. This affects how the sandwich pattern is applied:
+Opus 4.8 exhibits a distinct behavioral pattern: fewer tool calls with deeper reasoning per call. This affects how the sandwich pattern is applied:
 
-| Aspect | Opus 4.6 | Opus 4.7 |
+| Aspect | Opus 4.6 | Opus 4.8 |
 |--------|----------|----------|
 | Tool call frequency | Moderate | Lower |
 | Reasoning depth per call | Standard | Deeper |
 | Pre-reasoning benefit | Plan what to check | Plan explicit tool batch — model will under-call if not guided |
 
-### Pre-reasoning Adjustment for Opus 4.7
+### Pre-reasoning Adjustment for Opus 4.8
 
-When using Opus 4.7 in the pre-reasoning phase:
-1. **Explicitly enumerate tools needed** — Opus 4.7 may skip tool calls it deems unnecessary
+When using Opus 4.8 in the pre-reasoning phase:
+1. **Explicitly enumerate tools needed** — Opus 4.8 may skip tool calls it deems unnecessary
 2. **Batch tool-call plans upfront** — structure the action phase with a concrete tool sequence
 3. **Prefer fewer, richer tool calls** — align with the model's natural tendency
 
-Reference: [Best practices for using Claude Opus 4.7 with Claude Code](https://claude.com/blog/best-practices-for-using-claude-opus-4-7-with-claude-code)
+Reference: [Best practices for using Claude Opus 4.8 with Claude Code](https://claude.com/blog/best-practices-for-using-claude-opus-4-8-with-claude-code)

package/templates/.claude/skills/research/SKILL.md

@@ -194,15 +194,9 @@ Convergence expected by round 3. Hard stop at round 30.
 
 ### Tool: Writing artifacts under .claude/outputs/
 
-CC sensitive-path check inspects tool target paths and triggers permission prompts on `.claude/` regardless of `bypassPermissions` and allow rules (refs: #960, #961, #978, #981, #1016).
+Under `mode: "bypassPermissions"`, subagents write directly to `.claude/outputs/sessions/` with the Write tool — direct `.claude/**` writes are permitted (CC v2.1.121+, #1101). No `/tmp` staging or script wrapping is needed. Read-only Bash on `.claude/outputs/` (e.g., `cat`, `head`, `wc`) is allowed for verification.
 
-To write research results under `.claude/outputs/sessions/`:
-
-1. Write the artifact body to `/tmp/research-$(date +%H%M%S).md` first (Write tool target = `/tmp`, no sensitive-path trigger)
-2. Use a `/tmp/*.sh` Bash script to move/copy the file under `.claude/outputs/sessions/$(date +%Y-%m-%d)/` (Bash target = `/tmp`, script-internal `cp` to `.claude/` is not audited)
-3. Read-only Bash on `.claude/outputs/` (e.g., `cat`, `head`, `wc`) is allowed for verification
-
-Reference: `feedback_sensitive_path_tmp_bypass.md`, R006 sensitive-path handling, #1016, #1045.
+Reference: R006/R010 sensitive-path handling (direct `.claude/**` write under bypassPermissions), #1101.
 
    ```
    With metadata header:

package/templates/.claude/skills/result-aggregation/SKILL.md

@@ -170,15 +170,9 @@ R006 Artifact Channel Protocol을 소비하는 표준 패턴. 병렬 에이전
 
 ### Tool: Writing artifacts under .claude/outputs/
 
-CC sensitive-path check inspects tool target paths and triggers permission prompts on `.claude/` regardless of `bypassPermissions` and allow rules (refs: #960, #961, #978, #981, #1016).
+Under `mode: "bypassPermissions"`, subagents write directly to `.claude/outputs/sessions/` with the Write tool — direct `.claude/**` writes are permitted (CC v2.1.121+, #1101). No `/tmp` staging or script wrapping is needed. Read-only Bash on `.claude/outputs/` (e.g., `cat`, `head`, `wc`) is allowed for verification.
 
-To write result-aggregation results under `.claude/outputs/sessions/`:
-
-1. Write the artifact body to `/tmp/result-aggregation-$(date +%H%M%S).md` first (Write tool target = `/tmp`, no sensitive-path trigger)
-2. Use a `/tmp/*.sh` Bash script to move/copy the file under `.claude/outputs/sessions/$(date +%Y-%m-%d)/` (Bash target = `/tmp`, script-internal `cp` to `.claude/` is not audited)
-3. Read-only Bash on `.claude/outputs/` (e.g., `cat`, `head`, `wc`) is allowed for verification
-
-Reference: `feedback_sensitive_path_tmp_bypass.md`, R006 sensitive-path handling, #1016, #1045.
+Reference: R006/R010 sensitive-path handling (direct `.claude/**` write under bypassPermissions), #1101.
 
 
 ### 입력 형식

package/templates/.claude/skills/roundtable-debate/SKILL.md

@@ -50,14 +50,12 @@ TeamCreate("roundtable-{topic}")
 각 에이전트에게 위임 시 다음 directive를 prompt에 포함:
 
 ```
-**Sensitive-path artifact protocol (mandatory for this task)**
-
-If your task involves writing artifacts under `.claude/outputs/`:
-1. Build script in /tmp: cat > /tmp/roundtable-<persona>-<timestamp>.sh << 'EOF' ... EOF
-2. Script: mkdir -p .claude/outputs/sessions/<date>/ && cat > .claude/outputs/sessions/<date>/<artifact>.md << 'ARTIFACT' ... ARTIFACT
-3. Execute: bash /tmp/roundtable-<persona>-<timestamp>.sh
-4. Cleanup: rm /tmp/roundtable-<persona>-<timestamp>.sh
-DO NOT use Write/Edit directly on `.claude/outputs/` — CC sensitive-path guard triggers regardless of bypassPermissions/allow rules.
+**Artifact output (for this task)**
+
+If your task involves writing artifacts under `.claude/outputs/`, write the file to
+`.claude/outputs/sessions/<date>/<artifact>.md` using the Write tool directly (create the
+directory if needed). Your Agent tool call runs under `mode: "bypassPermissions"`, so direct
+Write/Edit on `.claude/` is permitted (CC v2.1.121+).
 ```
 
 ### Phase 1: Round 1 — Discussion

package/templates/.claude/skills/sauron-watch/SKILL.md

@@ -190,7 +190,7 @@ Starting full R017 verification...
 ═══════════════════════════════════════════════════════════
 
 [Round 1/5] mgr-supplier:audit
-  ✓ 41 agents checked, 0 issues
+  ✓ 49 agents checked, 0 issues
 
 [Round 2/5] mgr-updater:docs
   ✓ Documentation sync: OK

package/templates/.claude/skills/secretary-routing/SKILL.md

@@ -5,11 +5,11 @@ scope: core
 user-invocable: false
 context: fork
 ---
-## Mandatory delegation directive (R010 Universal /tmp Script Bypass)
+## Mandatory delegation directive (R010 bypassPermissions)
 
 When this skill spawns a subagent via the Agent tool, the spawned prompt MUST include this directive verbatim (or equivalent):
 
-> ALL tool calls targeting `.claude/` paths MUST go through `/tmp/*.sh` script via Bash. This includes read-only measurement (sed/cat/wc/ls/grep), Write, Edit. Read tool is exempt. Direct Write/Edit/Bash on .claude/ triggers user approval prompts that block unattended automation. See R010 for the full pattern.
+> 이 스킬이 Agent tool로 서브에이전트를 스폰할 때 mode: "bypassPermissions"를 반드시 포함한다. 서브에이전트는 .claude/ 경로에 직접 Write/Edit/Bash를 사용한다(/tmp 우회 불필요, CC v2.1.121+). See R010.
 
 This directive is preserved inline because Agent-tool prompt synthesis can drop SKILL.md notes; inline mandatory directives survive (#1046 lesson).
 

package/templates/.claude/skills/skill-extractor/SKILL.md

@@ -208,15 +208,9 @@ feedback memory에 누적된 실패 패턴을 분석하여 영구 구조(스킬
 
 ### Tool: Writing artifacts under .claude/outputs/
 
-CC sensitive-path check inspects tool target paths and triggers permission prompts on `.claude/` regardless of `bypassPermissions` and allow rules (refs: #960, #961, #978, #981, #1016).
+Under `mode: "bypassPermissions"`, subagents write directly to `.claude/outputs/sessions/` with the Write tool — direct `.claude/**` writes are permitted (CC v2.1.121+, #1101). No `/tmp` staging or script wrapping is needed. Read-only Bash on `.claude/outputs/` (e.g., `cat`, `head`, `wc`) is allowed for verification.
 
-To write skill-extractor results under `.claude/outputs/sessions/`:
-
-1. Write the artifact body to `/tmp/skill-extractor-$(date +%H%M%S).md` first (Write tool target = `/tmp`, no sensitive-path trigger)
-2. Use a `/tmp/*.sh` Bash script to move/copy the file under `.claude/outputs/sessions/$(date +%Y-%m-%d)/` (Bash target = `/tmp`, script-internal `cp` to `.claude/` is not audited)
-3. Read-only Bash on `.claude/outputs/` (e.g., `cat`, `head`, `wc`) is allowed for verification
-
-Reference: `feedback_sensitive_path_tmp_bypass.md`, R006 sensitive-path handling, #1016, #1045.
+Reference: R006/R010 sensitive-path handling (direct `.claude/**` write under bypassPermissions), #1101.
 
 
 ### 참조

package/templates/.claude/skills/spark-best-practices/SKILL.md

@@ -7,7 +7,7 @@ user-invocable: false
 
 # Apache Spark Best Practices
 
-> **Version**: Spark 4.0.2. Key changes from Spark 3.x: ANSI mode is now **default** (stricter SQL type coercion and overflow checks), and Spark Connect provides a decoupled client-server protocol for remote Spark access.
+> **Version**: Spark 4.x. Key changes from Spark 3.x: ANSI mode is now **default** (stricter SQL type coercion and overflow checks), and Spark Connect provides a decoupled client-server protocol for remote Spark access.
 
 ## Performance Optimization
 

package/templates/.claude/skills/status/SKILL.md

@@ -24,24 +24,22 @@ Show comprehensive system status including agents, skills, guides, and health ch
 AI Agent System - Status
 
 System:
-  Rules: 10 loaded (R000-R009)
+  Rules: 23 loaded (R000-R023)
 
 Agents:
-  Orchestrator:     1 (secretary)
-  Manager:          3 (mgr-creator, mgr-updater, mgr-supplier)
-  SW Engineer:      6
-  Backend Engineer: 3
+  Manager:          6 (mgr-creator, mgr-updater, mgr-supplier, ...)
+  SW Engineer:     21
+  DE Engineer:      6
+  Database:         4
   Infra Engineer:   2
-  Total:           15 agents
+  Other:           10 (security, architect, QA, system)
+  Total:           49 agents
 
 Skills:
-  Development:     8
-  Backend:         3
-  Infrastructure:  2
-  Total:          13 skills
+  Total:          118 skills
 
-Guides:            12 loaded
-Commands:          10 available
+Guides:            57 loaded
+Commands:          60+ available
 
 Health: ✓ OK
 ```
@@ -70,9 +68,6 @@ Rules:
     ✓ R009 parallel-execution
 
 Agents:
-  orchestrator/
-    ✓ secretary (internal)
-
   manager/
     ✓ mgr-creator (internal)
     ✓ mgr-updater (internal)
@@ -136,7 +131,7 @@ status --health
 Health Checks:
 
 Agents:
-  ✓ 15/15 agents valid
+  ✓ 49/49 agents valid
   ✓ All agent files exist in .claude/agents/
 
 Dependencies:

package/templates/.claude/skills/token-efficiency-audit/SKILL.md

@@ -125,7 +125,7 @@ Applies CI/Worker-Only Levers. These disable core oh-my-customcode functionality
 
 These settings disable core oh-my-customcode functionality:
   • CLAUDE_CODE_DISABLE_CLAUDE_MDS=1  → ALL rules and routing offline (R010 disabled)
-  • CLAUDE_AGENT_SDK_DISABLE_BUILTIN_AGENTS=1 → All 48 agents unavailable
+  • CLAUDE_AGENT_SDK_DISABLE_BUILTIN_AGENTS=1 → All 49 agents unavailable
   • ENABLE_CLAUDEAI_MCP_SERVERS=false → MCP-dependent skills unavailable
   • CLAUDE_CODE_DISABLE_AUTO_MEMORY=1 → No persistent memory across sessions
 

package/templates/.claude/skills/wiki/SKILL.md

@@ -1,5 +1,5 @@
 ---
-name: wiki
+name: omcustom:wiki
 description: Generate and maintain a persistent codebase wiki — LLM-built interlinked markdown knowledge base (Karpathy LLM Wiki pattern)
 scope: core
 version: 1.0.0

package/templates/.claude/skills/wiki-rag/SKILL.md

@@ -1,5 +1,5 @@
 ---
-name: wiki-rag
+name: omcustom:wiki-rag
 description: Use the project wiki as RAG knowledge source — search wiki pages to answer codebase questions before exploring raw files
 scope: core
 version: 1.0.0

package/templates/guides/airflow/README.md

@@ -1,10 +1,10 @@
-# Apache Airflow Guide (3.2.0)
+# Apache Airflow Guide (3.2.x)
 
-Reference documentation for Apache Airflow 3.2.0 DAG development best practices.
+Reference documentation for Apache Airflow 3.2.x DAG development best practices.
 
 ## Source
 
-Based on [Apache Airflow 3.2.0 official documentation](https://airflow.apache.org/docs/apache-airflow/3.2.0/) and [Astronomer best practices](https://docs.astronomer.io/).
+Based on [Apache Airflow 3.2.x official documentation](https://airflow.apache.org/docs/apache-airflow/stable/) and [Astronomer best practices](https://docs.astronomer.io/).
 
 ## Airflow 3.x Key Changes
 
@@ -39,9 +39,9 @@ This guide is referenced by:
 
 ## External Resources
 
-- [Airflow 3.2.0 Official Docs](https://airflow.apache.org/docs/apache-airflow/3.2.0/)
-- [Airflow 3.2.0 Best Practices](https://airflow.apache.org/docs/apache-airflow/3.2.0/best-practices.html)
-- [Airflow Task SDK](https://airflow.apache.org/docs/apache-airflow/3.2.0/authoring-and-scheduling/index.html)
-- [Airflow TaskFlow API](https://airflow.apache.org/docs/apache-airflow/3.2.0/core-concepts/taskflow.html)
-- [Migration Guide 2.x → 3.x](https://airflow.apache.org/docs/apache-airflow/3.2.0/migration-guide.html)
+- [Airflow 3.2.x Official Docs](https://airflow.apache.org/docs/apache-airflow/stable/)
+- [Airflow 3.2.x Best Practices](https://airflow.apache.org/docs/apache-airflow/stable/best-practices.html)
+- [Airflow Task SDK](https://airflow.apache.org/docs/apache-airflow/stable/authoring-and-scheduling/index.html)
+- [Airflow TaskFlow API](https://airflow.apache.org/docs/apache-airflow/stable/core-concepts/taskflow.html)
+- [Migration Guide 2.x → 3.x](https://airflow.apache.org/docs/apache-airflow/stable/migration-guide.html)
 - [Astronomer Docs](https://docs.astronomer.io/)

package/templates/guides/alembic/README.md

@@ -1,6 +1,6 @@
 # Alembic Guide
 
-Database migration framework for SQLAlchemy. Current stable: **1.18.x** (2025).
+Database migration framework for SQLAlchemy. Current stable: **1.18.x**.
 
 ## Overview
 

package/templates/guides/cc-token-saver/README.md

@@ -89,6 +89,20 @@ Shows per-session and cumulative cost. Useful for budget tracking across long se
 
 Reports your current rate limit hit to the community pool and shows aggregate rate limit data from other users. Helps gauge when limits reset.
 
+## Five-Layer Token Defense Stack
+
+cc-token-saver is **Layer 1 (Cache Defense)** of the token efficiency stack:
+
+- **Layer 1 — cc-token-saver** (this guide): Prompt cache TTL guard
+- **Layer 2 — R013 Ecomode**: Runtime behavior compression (see `.claude/rules/SHOULD-ecomode.md`)
+- **Layer 3 — Settings-based gates**: Pre-session config prevention via `/token-efficiency-audit` skill
+- **Layer 4 — playwright-compress**: Browser/tool output compression before it enters context
+- **Layer 5 — caveman**: English output token compression for code review and commit messages
+
+**Full stack overview**: `guides/claude-code/14-token-efficiency.md`
+
+The five layers are non-overlapping and independently deployable. Install cc-token-saver regardless of other layers.
+
 ## Integration Notes
 
 - R013 ecomode and Token Guardian are complementary, not competing

package/templates/guides/claude-code/03-tools.md

@@ -42,7 +42,7 @@ import anthropic
 client = anthropic.Anthropic()
 
 response = client.messages.create(
-    model="claude-sonnet-4-5",
+    model="claude-sonnet-5",
     max_tokens=1024,
     tools=[
         {

package/templates/guides/claude-code/06-mcp.md

@@ -26,7 +26,7 @@ import anthropic
 client = anthropic.Anthropic()
 
 response = client.beta.messages.create(
-    model="claude-sonnet-4-5",
+    model="claude-sonnet-5",
     max_tokens=1000,
     messages=[{
         "role": "user",

package/templates/guides/claude-code/11-sub-agents.md

@@ -12,14 +12,14 @@ Claude Code에서 특화된 서브에이전트를 생성하고 사용하는 방
 - **제약 적용**: 서브에이전트가 사용할 수 있는 도구 제한
 - **비용 제어**: 작업에 맞는 모델로 라우팅
 
-## Task Tool
+## Agent Tool (구 Task Tool)
 
-서브에이전트는 `Task` 도구를 통해 생성됩니다.
+서브에이전트는 `Agent` 도구(구 `Task` 도구)를 통해 생성됩니다.
 
 ### 기본 문법
 
 ```
-Task(
+Agent(
   subagent_type: "general-purpose",
   prompt: "수행할 작업 설명",
   model: "sonnet"
@@ -148,7 +148,7 @@ Task(prompt: "Task C", model: "haiku")
 
 ## 주의사항
 
-- 서브에이전트는 다른 서브에이전트를 생성할 수 없음
+- oh-my-customcode는 sole-orchestrator 설계 선택으로 서브에이전트가 서브에이전트를 스폰하지 않음 (CC v2.1.172+ 플랫폼은 5단계 중첩을 허용하나, oh-my-customcode는 예측 가능한 R009 병렬성/R018 조율을 위해 sole-orchestrator를 유지 — R010 참조)
 - 백그라운드 서브에이전트는 권한 프롬프트 자동 거부
 - 모델 미지정 시 기본값은 sonnet
 

package/templates/guides/claude-code/13-cli-flags.md

@@ -149,3 +149,4 @@ claude auto-mode critique    # Review auto mode decisions
 - [Monitoring Guide](10-monitoring.md) — OTel metrics and events
 - [Sub-agents Guide](11-sub-agents.md) — Agent spawning and coordination
 - [Workflow Patterns](12-workflow-patterns.md) — Workflow automation
+- `guides/claude-code/14-token-efficiency.md` — Five-layer token defense stack; this document's CLI flags feed into Layer 3 (settings-based gates)

package/templates/guides/claude-code/15-version-compatibility.md

@@ -1,11 +1,13 @@
 # Claude Code Version Compatibility
 
-> Updated: 2026-06-02
+> Updated: 2026-07-02
 > Source: Claude Code release notes (#967, #968, #969, #1126 auto-detected by claude-native skill, #1137, #1158, #1242, #1243, #1244, #1245, #1276, #1280)
+>
+> **Note (compat 노트 이관)**: 이 파일은 v2.1.159까지의 호환성 노트를 보존합니다. v2.1.160+ 이후 CC 호환성 노트는 `.claude/rules/` 각 규칙(R001/R002/R006/R010/R012 등)에 인라인으로 축적(이관)되었으므로, 최신 compat 노트는 해당 규칙 파일을 참조하세요.
 
 ## Compatibility Baseline
 
-oh-my-customcode v0.107.0 targets Claude Code v2.1.116+. All v2.1.117-119 additions are backward-compatible — no config changes required.
+oh-my-customcode v1.1.1 targets Claude Code v2.1.197+ (Sonnet 5 = CC 기본 모델). 이 파일에 정리된 v2.1.117-159 항목은 하위호환이며 config 변경이 불필요합니다.
 
 ## v2.1.117 (2026-04-22)
 

package/templates/guides/claude-code/index.yaml

@@ -50,6 +50,11 @@ documents:
     path: ./10-monitoring.md
     status: placeholder  # 공식 문서 미공개
 
+  - name: sub-agents
+    title: Sub-agent 가이드
+    description: Claude Code에서 특화된 서브에이전트를 생성하고 사용하는 방법
+    path: ./11-sub-agents.md
+
   - name: workflow-patterns
     title: 워크플로우 패턴
     description: AI 에이전트 작업에 적합한 워크플로우 패턴 선택 가이드
@@ -59,3 +64,13 @@ documents:
     title: CLI 플래그 및 모드
     description: CLI 플래그, 헤드리스 모드, 스케줄링, 환경변수 레퍼런스
     path: ./13-cli-flags.md
+
+  - name: token-efficiency
+    title: 토큰 효율성 — 5계층 방어 스택
+    description: cc-token-saver, Ecomode, settings 기반 게이트, playwright-compress, caveman의 조합
+    path: ./14-token-efficiency.md
+
+  - name: version-compatibility
+    title: Claude Code Version Compatibility
+    description: Claude Code 릴리즈 버전별 호환성 및 기능 변경 추적
+    path: ./15-version-compatibility.md

package/templates/guides/claude-code-tracking.md

@@ -8,7 +8,7 @@ oh-my-customcode가 의존하는 CC upstream API/기능 추적 목록. 각 항
 
 ### Status
 
-- **v2.1.146 기준**: 정식 progress API 미제공
+- **v2.1.197 기준**: 정식 progress API 미제공
 - `/bg` 흐름으로 시작한 background agent의 진행 상황을 메인 세션에서 직접 추적 불가
 
 ### Workarounds
@@ -38,7 +38,7 @@ oh-my-customcode가 의존하는 CC upstream API/기능 추적 목록. 각 항
 
 ### Status
 
-- **v2.1.146 기준**: `TeamDelete`에 `force` 옵션 미제공
+- **v2.1.197 기준**: `TeamDelete`에 `force` 옵션 미제공
 - Graceful shutdown 실패 시 tmux kill-pane workaround 필요
 
 ### Workaround

package/templates/guides/deep-plan/phases.md

@@ -126,19 +126,17 @@ Phase 1 research artifact is persisted by the `/research` skill.
 
 Phase 3 verification report path: `.claude/outputs/sessions/{YYYY-MM-DD}/deep-plan-{HHmmss}.md`
 
-**Sensitive-path artifact protocol**: Writing artifacts under `.claude/outputs/` MUST use the `/tmp/*.sh` bypass pattern. Direct `Write`/`Edit` and `Bash(mkdir -p)` on `.claude/` all trigger CC sensitive-path guard regardless of bypassPermissions. The `/tmp/*.sh` pattern lets the script internally write to `.claude/` — sensitive-path guard inspects only direct tool target paths, not script-internal file ops.
+**Artifact write protocol**: Under `mode: "bypassPermissions"`, write artifacts under `.claude/outputs/` directly with the Write tool (create the dated directory first if needed). Direct `Write`/`Edit`/`Bash` on `.claude/` is permitted since CC v2.1.121 — no `/tmp/*.sh` wrapping is needed (R006/R010). Only catastrophic shell operations (`rm -rf /`) remain blocked by independent safety guards.
 
 When spawning the final synthesis agent to write this artifact, include these instructions verbatim in the agent prompt:
 
 ```
-**Sensitive-path artifact protocol (mandatory for this task)**
-
-Write the artifact under `.claude/outputs/` using the /tmp/*.sh bypass:
-1. Build script in /tmp: cat > /tmp/deep-plan-<timestamp>.sh <<'EOF' ... EOF
-2. Script content: mkdir -p .claude/outputs/sessions/<date>/ && cat > .claude/outputs/sessions/<date>/deep-plan-<HHmmss>.md <<'ARTIFACT' ... ARTIFACT
-3. Execute: bash /tmp/deep-plan-<timestamp>.sh
-4. Cleanup: rm /tmp/deep-plan-<timestamp>.sh
-DO NOT use Write/Edit directly on `.claude/outputs/` — CC sensitive-path guard triggers regardless of bypassPermissions/allow rules.
+**Artifact write protocol (mandatory for this task)**
+
+Write the artifact under `.claude/outputs/` directly with the Write tool:
+1. Ensure the dated directory exists: .claude/outputs/sessions/<date>/
+2. Write .claude/outputs/sessions/<date>/deep-plan-<HHmmss>.md with the Write tool.
+This Agent tool call runs with mode: "bypassPermissions", under which direct Write/Edit/Bash on `.claude/` is permitted (CC v2.1.121+) — no /tmp/*.sh wrapping needed.
 ```
 
 Artifact metadata header:

package/templates/guides/external-tools/ecc-absorption-decisions.md

@@ -206,7 +206,7 @@ R016 부담 없이 80% 이상의 use case를 커버합니다.
 
 DEFER 조건 모니터링을 위한 내부 KPI 메모:
 
-> 내부 메모: `.claude/agent-memory/sys-memory-keeper/project_ecc_kpi_internal_metrics.md`
+> 내부 메모: sys-memory-keeper 프로젝트 메모리에 KPI 추적 (예정 — 파일 미생성)
 
 | KPI | 6개월 목표 | 1년 목표 |
 |-----|-----------|---------|
@@ -228,5 +228,5 @@ KPI 초과 달성 시: R016 워크플로우로 재검토 트리거.
 | sec-agentshield-wrapper | #1174 |
 | instinct-extractor | #1175 |
 | manifest-install --profile | #1177 |
-| 시퀀싱 메모 | `.claude/agent-memory/sys-memory-keeper/` → `[[project-sequencing-alpha-beta-gamma]]` |
+| 시퀀싱 메모 | sys-memory-keeper 프로젝트 메모리 (예정 — 파일 미생성) |
 | R006 compilation metaphor | `.claude/rules/MUST-agent-design.md` |

package/templates/guides/external-tools/graphify-integration.md

@@ -83,6 +83,6 @@ source:
 
 - `.claude/skills/wiki/SKILL.md` — 위키 생성/유지 스킬 (Karpathy LLM Wiki 패턴)
 - `.claude/skills/wiki-rag/SKILL.md` — wiki/index.yaml 기반 RAG 조회 스킬
-- `.claude/skills/ontology-rag/SKILL.md` — ontology-rag MCP 기반 라우팅 enrichment
+- ontology-rag MCP 서버 (R019 참조) — 라우팅 enrichment Layer-1 (`.claude/rules/SHOULD-ontology-rag-routing.md`)
 - `.claude/rules/SHOULD-ontology-rag-routing.md` — R019 dual-layer enrichment 규칙
 - 내재화 배경: issue #977

package/templates/guides/git-worktree-workflow/README.md

@@ -72,7 +72,7 @@ Claude Code provides `EnterWorktree` and `ExitWorktree` tools for session-scoped
 
 ```
 EnterWorktree(name: "feature-x")
-# Creates .claude/worktrees/feature-x with a new branch based on HEAD
+# Creates .claude/worktrees/feature-x with a new branch; base ref governed by the worktree.baseRef setting (fresh = origin/<default-branch>, the default; head = branch from current HEAD, opt-in)
 # Session working directory switches to the worktree
 
 EnterWorktree(path: "/absolute/path/to/existing-worktree")

package/templates/guides/index.yaml

@@ -55,7 +55,7 @@ guides:
     source:
       type: external
       origin: docs.oracle.com
-      url: https://docs.oracle.com/en/java/javase/21/
+      url: https://docs.oracle.com/en/java/javase/25/
 
   - name: python
     description: Python reference from PEP 8 and PEP 20

package/templates/guides/multi-model-routing/README.md

@@ -9,11 +9,16 @@ Role-based model selection strategy for AI agent workflows. Consolidates model r
 | Alias | Full ID | Cost | Speed | Use Case |
 |-------|---------|------|-------|----------|
 | `haiku` | claude-haiku-4-5 | $ | Fast | Search, simple edits, file discovery |
-| `sonnet` | claude-sonnet-4-6 | $$ | Moderate | Code generation, general tasks (default) |
+| `sonnet` | claude-sonnet-4-6 | $$ | Moderate | Code generation, general tasks (project base alias — pinned to 4-6 for stability) |
+| `sonnet5` | claude-sonnet-5 | $$ | Moderate | CC platform default (v2.1.197+); native 1M context |
 | `opus` | claude-opus-4-6 | $$$ | Slower | Complex reasoning, architecture, planning |
+| `opus48` | claude-opus-4-8 | $$$ | Slower | Latest Opus (GA); highest capability below Fable 5 |
+| `fable` | claude-fable-5 | $$$$ | Slower | Mythos-class; tier above Opus, highest GA capability |
 | `opusplan` | claude-opus-4-6 + plan mode | $$$ | Slower | Architecture with approval gates |
 
-Extended context: `[1m]` suffix enables 1M token context (e.g., `claude-opus-4-6[1m]`).
+Extended context: `[1m]` suffix enables 1M token context (e.g., `claude-opus-4-6[1m]`). Fable 5 and Sonnet 5 include 1M context by default — no `[1m]` suffix needed.
+
+> **Base alias pin (design intent)**: The base `sonnet` alias stays pinned to `claude-sonnet-4-6` for stability — existing agents specifying `model: sonnet` remain unchanged. Claude Sonnet 5 (`claude-sonnet-5`) is the CC platform default as of v2.1.197 and is opt-in via the explicit `sonnet5` alias.
 
 ## Role-Based Routing Table
 

package/templates/guides/professor-triage/README.md

@@ -37,6 +37,6 @@ The SKILL.md is intentionally thin — it carries only the workflow contract and
 
 ## Key Design Decisions
 
-- **Phase 4 uses `general-purpose` (NOT `arch-documenter`)**: `arch-documenter` has `disallowedTools: [Bash]` — cannot execute `/tmp/*.sh` bypass → falls back to Write → triggers sensitive-path guard. `general-purpose` has Bash access. See #1043.
-- **Sensitive-path protocol**: All `.claude/outputs/` writes must use `/tmp/*.sh` bypass. See phases.md Phase 4E.
+- **Phase 4 uses `general-purpose` (NOT `arch-documenter`)**: `arch-documenter` has `disallowedTools: [Bash]` — cannot run `gh` (comment posting) or shell scripts these phases require. `general-purpose` has Bash access. See #1043 and R010 Known Limitations.
+- **Artifact write**: `.claude/outputs/` writes use the Write tool directly under `mode: "bypassPermissions"` — direct Write/Edit/Bash on `.claude/` is permitted since CC v2.1.121 (R006/R010); no `/tmp/*.sh` wrapping needed. See phases.md Phase 4E.
 - **Parallelization**: Phase 4A + 4B parallel; Phase 4C after both; Phase 4D + 4E parallel; Phase 4F verification gate.