oh-my-customcode 0.154.1 → 0.156.0

package/dist/cli/index.js

@@ -2334,7 +2334,7 @@ var init_package = __esm(() => {
     workspaces: [
       "packages/*"
     ],
-    version: "0.154.1",
+    version: "0.156.0",
     description: "Batteries-included agent harness for Claude Code",
     type: "module",
     bin: {
@@ -2382,7 +2382,7 @@ var init_package = __esm(() => {
       yaml: "^2.8.2"
     },
     devDependencies: {
-      "@anthropic-ai/sdk": "^0.97.1",
+      "@anthropic-ai/sdk": "^0.98.0",
       "@biomejs/biome": "^2.3.12",
       "@types/bun": "^1.3.6",
       "@types/js-yaml": "^4.0.9",

package/dist/index.js

@@ -2031,7 +2031,7 @@ var package_default = {
   workspaces: [
     "packages/*"
   ],
-  version: "0.154.1",
+  version: "0.156.0",
   description: "Batteries-included agent harness for Claude Code",
   type: "module",
   bin: {
@@ -2079,7 +2079,7 @@ var package_default = {
     yaml: "^2.8.2"
   },
   devDependencies: {
-    "@anthropic-ai/sdk": "^0.97.1",
+    "@anthropic-ai/sdk": "^0.98.0",
     "@biomejs/biome": "^2.3.12",
     "@types/bun": "^1.3.6",
     "@types/js-yaml": "^4.0.9",

package/package.json

@@ -3,7 +3,7 @@
   "workspaces": [
     "packages/*"
   ],
-  "version": "0.154.1",
+  "version": "0.156.0",
   "description": "Batteries-included agent harness for Claude Code",
   "type": "module",
   "bin": {
@@ -51,7 +51,7 @@
     "yaml": "^2.8.2"
   },
   "devDependencies": {
-    "@anthropic-ai/sdk": "^0.97.1",
+    "@anthropic-ai/sdk": "^0.98.0",
     "@biomejs/biome": "^2.3.12",
     "@types/bun": "^1.3.6",
     "@types/js-yaml": "^4.0.9",

package/templates/.claude/rules/MUST-continuous-improvement.md

@@ -42,6 +42,8 @@ When CI failure, process gap, or repeatable system defect is found:
 2. Register GitHub issue (trackable improvement item)
 3. Both required — memory alone is NOT sufficient for system-level defects
 
+> **Auto-register directive (session 102)**: Genuine defects and process gaps surfaced during release workflows (e.g., `post-release-followup` skill) MUST be registered as GitHub issues automatically — no user confirmation required. Only code-changing "immediate action" items require user confirmation. See `post-release-followup` skill: "Auto-Register Genuine Defects (no-ask)" section.
+
 ### Adaptive Harness Integration
 
 When repeating agent failures or suboptimal routing is detected:

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

@@ -114,15 +114,7 @@ When ALL reviewers agree BUILD or BUILD WITH CHANGES:
 
 ### 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).
-
-To write agora results under `.claude/outputs/sessions/`:
-
-1. Write the artifact body to `/tmp/agora-$(date +%H%M%S).md` first (Write tool target = `/tmp`, no sensitive-path trigger)
-2. Use a `/tmp/*.sh` Bash script to move/copy the file under `.claude/outputs/sessions/$(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.
+Direct Write to `.claude/outputs/sessions/{date}/agora-{topic}-{time}.md` is permitted under `mode: "bypassPermissions"` (CC v2.1.121+, #1101). Write tool auto-creates parent directory. For CC < v2.1.121, see git history for the legacy /tmp bypass pattern.
 
 3. Shut down team: `SendMessage(to: "*", message: {type: "shutdown_request"})`
 

package/templates/.claude/skills/claude-native/SKILL.md

@@ -42,18 +42,18 @@ Search for existing tracking issues to avoid duplicates:
 ```bash
 gh issue list \
   --state all \
-  --search "[Claude Code v" \
+  --search "Claude Code v" \
   --json number,title \
   --limit 100
 ```
 
-Build a set of already-tracked versions by extracting version strings from issue titles matching the pattern `[Claude Code v{version}]`.
+Build a set of already-tracked versions by extracting version strings from issue titles matching the pattern `Claude Code v(\d+\.\d+\.\d+)` (no brackets).
 
 ### Phase 3: Dedup
 
 For each fetched release version:
 - Parse the version string from `tag_name` (e.g., `v2.1.86`)
-- If a matching issue title already exists → skip (already tracked)
+- If an issue title matching `Claude Code v{version}` already exists → skip (already tracked)
 - If no matching issue → add to "needs issue" list
 
 ### Phase 4: Create Issues (or Dry-Run Report)
@@ -76,7 +76,7 @@ For each version in the "needs issue" list, create a GitHub issue:
 
 ```bash
 gh issue create \
-  --title "[Claude Code v{version}] New release detected" \
+  --title "Claude Code v{version}" \
   --label "automated,claude-code-release" \
   --body "{body}"
 ```
@@ -105,7 +105,7 @@ Issue body format (matching the pattern established by issue #683):
 
 ---
 
-_This issue was created by the `/omcustom:claude-native` skill._
+_This issue was auto-created by the cc-release-monitor workflow (claude-native skill)._
 ```
 
 **Notes:**
@@ -128,8 +128,8 @@ Versions checked: {N}
 New issues created: {M}
 
 Created:
-  - #1234 [Claude Code v2.1.86] New release detected
-  - #1235 [Claude Code v2.1.87] New release detected
+  - #1234 Claude Code v2.1.86
+  - #1235 Claude Code v2.1.87
 
 Already tracked (skipped):
   - v2.1.85 → #683
@@ -157,6 +157,8 @@ For each release:
 
 Semver comparison: major → minor → patch (all numeric). Pre-release suffixes (e.g., `-beta`) are included and compared lexicographically after numeric parts.
 
+**Note on non-contiguous patch numbers**: Claude Code skips some patch numbers (e.g., v2.1.151 and v2.1.155 were never released publicly). The skill MUST act only on versions that actually appear in the GitHub releases API response — never assume contiguous numbering or attempt to fill gaps.
+
 ## Error Handling
 
 | Error | Action |

package/templates/.claude/skills/codex-exec/SKILL.md

@@ -252,15 +252,7 @@ codex-exec "build/fix frontend"
 
 ### 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).
-
-To write codex execution results under `.claude/outputs/codex/`:
-
-1. Write the artifact body to `/tmp/codex-{HHmmss}.{ext}` first (Write tool target = /tmp, no sensitive-path trigger)
-2. Use a `/tmp/*.sh` Bash script to move/copy the file under `.claude/outputs/codex/sessions/...` (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.
+Direct Write to `.claude/outputs/codex/sessions/{date}/codex-{HHmmss}.{ext}` is permitted under `mode: "bypassPermissions"` (CC v2.1.121+, #1101). Write tool auto-creates parent directory. For CC < v2.1.121, see git history for the legacy /tmp bypass pattern.
 
 ### Attribution
 

package/templates/.claude/skills/hada-scout/SKILL.md

@@ -1,92 +1,144 @@
 ---
 name: hada-scout
-description: hada.io RSS feed monitoring for AI agent/harness articles with automated /scout analysis
-scope: package
-version: 1.0.0
-user-invocable: false
+description: hada.io RSS feed monitoring with LLM pre-scout filtering for oh-my-customcode relevance
+scope: core
+version: 2.0.0
+user-invocable: true
+argument-hint: "[--limit N] [--threshold N]"
 ---
 
 # hada-scout
 
-Automated pipeline that monitors hada.io (via feedburner RSS) for AI agent, harness, benchmark, and eval-related articles, then runs `/scout` analysis on each match.
+3-phase in-session pipeline that monitors hada.io (via feedburner RSS) for relevant articles,
+uses a haiku LLM batch to pre-score relevance, and dispatches full `/scout` analysis only on
+high-scoring candidates.
 
 ## Purpose
 
-Complement geeknews-scout with harness/eval-focused coverage from hada.io. While geeknews-scout casts a broad net over AI agent news, hada-scout narrows to benchmark/evaluation framework content — the domain most relevant to oh-my-customcode's harness and agent-eval subsystems.
+Replace the v1.0 keyword-regex approach with context-aware LLM pre-scoring. Haiku evaluates
+all feed titles in a single batch call, reducing false positives from ~30-40% to ~5-10% and
+eliminating the external CronJob dependency for user-invoked runs.
 
-## Architecture: 2-Layer Hybrid
+## Architecture: 3-Phase In-Session Pipeline
 
-### Layer 1 — check-feed.sh (feed → issues)
+### Phase 1 — Fetch & Parse
 
-1. Fetch hada.io feedburner RSS
-2. Filter entries by keyword regex (case-insensitive)
-3. Dedup against existing `hada-scout` issues
-4. Create GitHub issue per match with labels `hada-scout` + `pending-scout`
+1. WebFetch the hada.io RSS feed (`https://feeds.feedburner.com/geeknews-feed`)
+2. Parse all items: title, URL, publication date
+3. Default: latest 50 items (configurable via `--limit` or `HADA_SCOUT_LIMIT`)
 
-### Layer 2 — scout-runner.sh (issues → /scout)
+### Phase 2 — Pre-Scout (haiku batch)
 
-1. Find open issues with `pending-scout` label
-2. Extract source URL from issue body
-3. Run `claude -p "/scout {url}"` (max 5 executions per run)
-4. Parse verdict from /scout output
-5. Apply verdict label, remove `pending-scout`
+1. Spawn 1 haiku agent with ALL item titles as a single batch input
+2. Agent evaluates each title against oh-my-customcode's domain (see prompt template below)
+3. Returns relevance score (0–100) and a 1-line reason for each item
+4. Threshold: ≥ 60 passes to Phase 3 (configurable via `--threshold` or `HADA_SCOUT_THRESHOLD`)
+5. Cost: ~$0.01–0.05 for 50 items
 
-## Keyword Strategy
+### Phase 3 — Scout Dispatch
 
-hada-scout uses harness/benchmark/eval focused keywords, distinct from geeknews-scout's broader AI agent coverage:
+1. Dedup: check existing `hada-scout` labeled issues via `gh issue list --label hada-scout`
+2. For each passing item (max 5 per run, configurable via `MAX_SCOUT_PER_RUN`):
+   - Run full `/scout` analysis via Skill tool invocation
+   - Scout creates GitHub issue with verdict label on `baekenough/oh-my-customcode`
+   - Add `hada-scout` label to the created issue
+3. Dispatch up to 4 scouts in parallel per R009
+
+## Pre-Scout Prompt Template
+
+The haiku agent receives the following system prompt:
 
 ```
-harness|benchmark|eval|evaluation framework|agent framework|코드 리뷰 자동화|하네스|벤치마크|평가
+You are a relevance filter for the oh-my-customcode project — an AI agent harness/orchestration
+system built on Claude Code CLI with 44 agents, 74 skills.
+
+Project domains (HIGH relevance):
+- AI agent orchestration, multi-agent systems, agent design patterns
+- Harness, benchmark, evaluation frameworks for AI agents
+- Claude Code, Anthropic ecosystem, MCP protocol
+- Code review automation, development workflow automation
+- Agent sandbox, isolation, security patterns
+- LLM-assisted development tools and methodologies
+
+Project domains (MEDIUM relevance):
+- General AI/ML tooling that could be adapted for agent workflows
+- DevOps automation patterns applicable to agent infrastructure
+- New programming paradigms for AI-assisted development
+
+NOT relevant:
+- Pure frontend/UI frameworks without agent connection
+- Business/management topics
+- Hardware, networking, non-AI infrastructure
+- Social media, marketing tools
+
+For each item below, return: score (0-100) | reason (1 line)
+
+Items:
+{numbered_item_list}
 ```
 
-Geeknews-scout handles: `Claude|Anthropic|MCP|AI agent|에이전트|agentic|multi-agent|...`
+## Display Format
+
+```
+[hada-scout] Scanning hada.io feed...
+├── Phase 1: Fetched {n} items
+├── Phase 2: Pre-scout → {passed}/{total} items passed (threshold: {t}%)
+│   ├── ✓ {title1} (score: {s1}%)
+│   ├── ✓ {title2} (score: {s2}%)
+│   └── ✗ {title3} (score: {s3}%) — skipped
+├── Phase 3: Scout dispatch ({n} items, max 5)
+│   ├── [1] /scout {url1} → {verdict}
+│   └── [2] /scout {url2} → {verdict}
+└── [Done] {created}/{dispatched} issues created
+```
 
 ## Label Scheme
 
 | Label | Purpose |
 |-------|---------|
 | `hada-scout` | Source identification — all hada-scout created issues |
-| `pending-scout` | Awaiting /scout analysis (set by Layer 1, cleared by Layer 2) |
 | `scout:internalize` | /scout verdict: adopt into project |
 | `scout:integrate` | /scout verdict: use as external dependency |
 | `scout:skip` | /scout verdict: not relevant |
 
 ## Cost Controls
 
-- Layer 2 runs at most **5 /scout executions per cron invocation**
-- Each /scout call costs ~$0.5–1.5 (sonnet)
-- Remaining `pending-scout` issues are processed in the next scheduled run
-
-## Deployment
-
-- Pattern: same K8s CronJob structure as `infra/geeknews-scout/`
-- Host: `ubuntu-ext` cluster
-- Infrastructure files: `infra/hada-scout/`
-  - `check-feed.sh` — Layer 1 feed poller
-  - `scout-runner.sh` — Layer 2 /scout executor
-  - `Dockerfile`
-  - `cronjob.template.yaml`
-  - `deploy.sh`
-  - `.env.example`
+| Stage | Model | Estimated Cost |
+|-------|-------|----------------|
+| Pre-scout (Phase 2) | haiku | ~$0.01–0.05 per run (50 items) |
+| Full scout (Phase 3) | sonnet | ~$0.5–1.5 per item, max 5 per run |
+| Total max per invocation | — | ~$8 |
 
 ## Environment Variables
 
 | Variable | Default | Description |
 |----------|---------|-------------|
-| `GH_TOKEN` | (required) | GitHub PAT for issue creation |
-| `REPO` | `baekenough/oh-my-customcode` | Target repo |
-| `FEED_URL` | `http://feeds.feedburner.com/geeknews-feed` | hada.io RSS feed |
-| `KEYWORDS` | (see above) | Pipe-separated keyword regex |
-| `MAX_SCOUT_PER_RUN` | `5` | Max /scout executions per Layer 2 run |
+| `FEED_URL` | `https://feeds.feedburner.com/geeknews-feed` | RSS feed URL |
+| `HADA_SCOUT_THRESHOLD` | `60` | Pre-scout score threshold (0–100) |
+| `HADA_SCOUT_LIMIT` | `50` | Max feed items to fetch and score |
+| `MAX_SCOUT_PER_RUN` | `5` | Max /scout executions per invocation |
+| `GH_TOKEN` | (required) | GitHub PAT for issue creation and dedup |
 
 ## Integration
 
 | Rule | How |
 |------|-----|
-| R009 | Layer 1 and Layer 2 run as independent CronJobs |
-| R010 | Skill defines the architecture; implementation delegated to infra-docker-expert for K8s manifests |
-| scout skill | Layer 2 invokes `/scout` via `claude -p` subprocess |
-
-## Tracking Issue
+| R009 | Phase 3 scout dispatches run in parallel (up to 4 concurrent) |
+| R010 | Orchestrator manages phases; analysis delegated to haiku/sonnet agents |
+| R015 | Pre-scout scores and reasons displayed before dispatching full scouts |
+| scout skill | Phase 3 invokes `/scout` via Skill tool for each candidate URL |
+
+## Differences from v1.0
+
+| Aspect | v1.0 (keyword) | v2.0 (LLM pre-scout) |
+|--------|----------------|----------------------|
+| Filtering | Regex keyword match | LLM relevance scoring (haiku) |
+| Invocation | External CronJob only | User-invocable `/hada-scout` + CronJob |
+| Precision | Low (keyword false positives) | High (context-aware scoring) |
+| Cost per scan | $0 (regex) + $2.5–7.5 (/scout) | $0.05 (pre-scout) + $2.5–7.5 (/scout) |
+| False positive rate | ~30–40% | ~5–10% |
+| Scope | `package` | `core` |
+
+## Tracking
 
 GitHub Issue #841

package/templates/.claude/skills/memory-management/SKILL.md

@@ -194,3 +194,47 @@ recall_errors:
   - Connection failure: Return empty with warning
   - Invalid query: Help user reformulate
 ```
+
+## MemKraft Bridge (Optional)
+
+> External integration: [MemKraft](https://github.com/seojoonkim/memkraft) — zero-dependency compound memory for AI agents.
+> Install: `pipx install memkraft`
+
+### When to Use
+
+| Capability | claude-mem | MemKraft |
+|-----------|-----------|----------|
+| Session persistence | ✅ (Chroma) | ✅ (Markdown) |
+| Entity tracking | ❌ | ✅ (person/org/concept) |
+| Source attribution | ❌ | ✅ (`[Source: who, when, how]`) |
+| Auto-maintenance | ❌ | ✅ (Dream Cycle) |
+| CJK entity extraction | ❌ | ✅ (Korean/Chinese/Japanese) |
+| Offline search | ❌ | ✅ (stdlib difflib) |
+
+Use MemKraft when entity tracking or source attribution is needed. Use claude-mem for simple session persistence.
+
+### Commands
+
+```bash
+# Extract entities from a document
+memkraft extract <file>
+
+# Get a brief on a topic
+memkraft brief <topic>
+
+# Run maintenance cycle (dedup, prune orphans)
+memkraft dream
+```
+
+### Integration with sys-memory-keeper
+
+At session end, sys-memory-keeper can optionally run MemKraft operations:
+
+1. `memkraft extract` on session summary → builds entity graph
+2. `memkraft dream` → prunes stale entries (run weekly, not every session)
+
+### Prerequisites
+
+- Python 3.9+
+- `pipx install memkraft`
+- No API keys required (offline-only)

package/templates/.claude/skills/omcustom-auto-improve/SKILL.md

@@ -66,8 +66,8 @@ Map each approved item to the appropriate subagent by `targetType`:
 | targetType | Agent | Action |
 |------------|-------|--------|
 | agent | mgr-creator | Modify agent frontmatter/body |
-| skill | Matching domain expert | Revise skill SKILL.md |
-| routing | general-purpose | Update routing patterns |
+| skill | mgr-creator | Revise skill SKILL.md (R010 Protected Paths) |
+| routing | mgr-creator | Update routing patterns (R010 Protected Paths) |
 | model-escalation | general-purpose | Update model field in agent frontmatter |
 
 Spawn agents in parallel (max 4 per R009). Each agent receives:

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

@@ -63,6 +63,8 @@ Execute these steps to display available pipelines:
    - **Skill steps** (`skill: name`): Invoke via Skill tool — `Skill(skill: "{name}")`
    - **Prompt steps** (`prompt: text`): Execute the described action using appropriate agents/tools
    - **Foreach steps** (`foreach: collection`): Iterate over collection from previous step output
+   - **Parallel steps** (`parallel: [step1, step2]`): Execute contained steps concurrently using Agent tool. Each parallel step runs as an independent Agent. Max 4 concurrent per R009. Steps within a parallel block MUST be independent (no shared state, no sequential dependencies). Dependencies between parallel and non-parallel steps use `depends_on:` field.
+   - **Agent mode**: When spawning agents via Agent tool, always pass `mode: "bypassPermissions"` to match agent frontmatter. The Agent tool default (`acceptEdits`) overrides frontmatter `permissionMode`, causing permission prompts during unattended execution.
 5. Report completion or failure
 
 ### Resume Mode (/pipeline resume)
@@ -94,6 +96,48 @@ Track per-step state:
 
 State saved to `/tmp/.claude-pipeline-{name}-{PPID}.json` on failure.
 
+## Parallel Execution
+
+Pipeline steps can be grouped for parallel execution:
+
+```yaml
+steps:
+  - name: phase-1
+    parallel:
+      - name: task-a
+        skill: skill-a
+        description: First independent task
+      - name: task-b
+        skill: skill-b
+        description: Second independent task
+  - name: phase-2
+    skill: next-step
+    depends_on: phase-1
+```
+
+### Parallel Rules
+
+- Max 4 concurrent steps per parallel block (R009 hard cap)
+- Steps within a parallel block MUST be independent
+- `depends_on` enforces ordering between blocks
+- Each parallel step is spawned as a separate Agent tool call in the SAME message
+- If any parallel step fails with `error: halt-and-report`, all remaining steps in the block are cancelled
+- State tracking records each parallel step individually
+
+### Parallel State Format
+
+```json
+{
+  "name": "phase-1",
+  "type": "parallel",
+  "status": "running",
+  "children": [
+    {"name": "task-a", "status": "completed", "duration_ms": 5000},
+    {"name": "task-b", "status": "running"}
+  ]
+}
+```
+
 ## Error Handling
 
 - Pipeline not found → list available pipelines with suggestion

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

@@ -10,7 +10,7 @@ effort: medium
 
 ## Purpose
 
-After PR creation in the auto-dev release workflow, collect unaddressed findings and present actionable follow-up recommendations. The user chooses: execute now, register as issues, or skip.
+After PR creation in the auto-dev release workflow, collect unaddressed findings and present actionable follow-up recommendations. Genuine defects and process gaps are registered as GitHub issues automatically (no user confirmation needed). Only code-changing "immediate action" items require user confirmation.
 
 ## Workflow
 
@@ -63,55 +63,73 @@ Remove duplicates (same issue referenced from multiple sources). Categorize:
 | **이슈 등록** | P3 이슈, LOW 검증 발견사항, 새 TODO, Medium PR 리뷰 발견사항 | 이슈로 등록 |
 | **참고** | 이미 추적 중인 이슈, 외관 관련 메모 | 건너뛰기 |
 
+### Auto-Register Genuine Defects (no-ask)
+
+**Before presenting the summary to the user**, auto-register all "이슈 등록" category items that are genuine defects or process gaps. No user confirmation is required for these.
+
+**Definition — auto-register if ANY of:**
+- Genuine defect: a bug, regression, broken behavior, or incorrect output observed during verification
+- Process gap: a workflow hole, missing guard, or coverage gap surfaced by deep-verify / triage
+- Coverage gap: a missing test, missing documentation, or missing automation for a known scenario
+
+**Definition — do NOT auto-register (keep for user choice):**
+- Pure cosmetic / style preference notes
+- Items that are purely subjective or opinion-based
+
+**When ambiguous, lean toward registering.** Registering a borderline item costs nothing; missing a genuine defect costs a future session.
+
+**How to auto-register:**
+```bash
+gh issue create \
+  --title "{간결한 설명}" \
+  --body "## 출처\n\nv{version} 릴리즈 워크플로우에서 자동 등록.\n\n## 컨텍스트\n\n{상세 컨텍스트}\n\n## 권장 조치\n\n{권장 사항}" \
+  --label "professor"
+```
+Add priority label (`P3` default for defects surfaced here; escalate to `P2` if MEDIUM+ severity).
+
+**Authority**: user directive (session 102) — genuine defects found during release workflows should be registered without prompting. See also R016 Defect Response Matrix: CI/infra defect and Process gap both require Issue registration.
+
+After auto-registering, include the created issue numbers in the summary display.
+
 ### 3. Present to User
 
-Display follow-up summary:
+Auto-register all genuine defects first (see above). Then display follow-up summary showing what was already registered and what remains for user decision:
 
 ```
 [Follow-up] {n}개 후속 작업 발견
 
+━━━ 자동 등록 완료 ({count}개) ━━━
+  ✓ #{issue_number} — {description} (이미 등록됨)
+  ✓ #{issue_number} — {description} (이미 등록됨)
+
 ━━━ 즉시 실행 추천 ({count}개) ━━━
   1. {description} — 출처: {source}
   2. {description} — 출처: {source}
 
-━━━ 이슈 등록 추천 ({count}개) ━━━
-  3. {description} — 출처: {source}
-  4. {description} — 출처: {source}
-
 ━━━ 참고 사항 ({count}개) ━━━
-  5. {description} — 이미 #{issue_number}로 추적 중
-
-선택:
-  [A] 추천대로 실행 (즉시 실행 + 이슈 등록)
-  [B] 모두 즉시 실행
-  [C] 모두 이슈 등록
-  [D] 개별 선택 (항목별로 질문)
-  [E] 건너뛰기
+  3. {description} — 이미 #{issue_number}로 추적 중
+
+즉시 실행 항목 선택:
+  [A] 추천대로 실행 (즉시 실행 항목 모두 실행)
+  [B] 개별 선택 (항목별로 질문)
+  [C] 건너뛰기
 ```
 
-Use AskUserQuestion (or equivalent user prompt) to get the choice.
+Use AskUserQuestion (or equivalent user prompt) to get the choice **only if there are "즉시 실행" items**. If there are none, skip the prompt and complete automatically.
 
 ### 4. Process User Choice
 
 **Option A (추천대로)**:
 - "Immediate" items → delegate to appropriate specialist agents for execution
-- "Trackable" items → create GitHub issues via `gh issue create`
+- (Trackable items were already auto-registered in step above)
 - "Informational" items → skip
 
-**Option B (모두 즉시 실행)**:
-- All Immediate + Trackable items → delegate to specialist agents
-- Follow implementation patterns from the release workflow
-
-**Option C (모두 이슈 등록)**:
-- All Immediate + Trackable items → `gh issue create` with appropriate labels
-- Label: `professor` for auto-triage in next workflow run
-
-**Option D (개별 선택)**:
-- For each item, ask: `[{n}] {description} — 실행(E) / 이슈(I) / 건너뛰기(S)?`
+**Option B (개별 선택)**:
+- For each "즉시 실행" item, ask: `[{n}] {description} — 실행(E) / 건너뛰기(S)?`
 - Process each per user choice
 
-**Option E (건너뛰기)**:
-- Skip all follow-up actions
+**Option C (건너뛰기)**:
+- Skip remaining immediate actions
 - Complete workflow
 
 ### 5. Report
@@ -126,7 +144,16 @@ Use AskUserQuestion (or equivalent user prompt) to get the choice.
 
 ## Issue Creation Template
 
-When creating follow-up issues:
+For auto-registered genuine defects / process gaps:
+
+```bash
+gh issue create \
+  --title "{간결한 설명}" \
+  --body "## 출처\n\nv{version} 릴리즈 워크플로우에서 자동 등록.\n\n## 컨텍스트\n\n{triage/verify에서의 상세 컨텍스트}\n\n## 권장 조치\n\n{권장 사항}" \
+  --label "professor"
+```
+
+For user-requested issue creation (Option C fallback, if ever used):
 
 ```bash
 gh issue create \
@@ -135,12 +162,13 @@ gh issue create \
   --label "professor"
 ```
 
-Add priority label (`P1`, `P2`, `P3`) based on categorization.
+Add priority label (`P1`, `P2`, `P3`) based on categorization. Default for auto-registered items: `P3` (escalate to `P2` for MEDIUM+ severity).
 
 ## Notes
 
 - This skill runs in the main conversation context (via workflow skill step)
-- User interaction is expected — this is NOT a fully automated step
+- Genuine defect/process gap items are auto-registered as issues WITHOUT user confirmation (session 102 directive)
+- Only "즉시 실행" (code-changing immediate action) items require user confirmation
 - All file modifications delegated to specialist subagents per R010
 - Issue creation uses `gh` CLI directly (read-only operation pattern)
 - If no follow-up candidates found, report "No follow-up actions needed" and complete

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

@@ -22,17 +22,8 @@ Switch the active plugin set to match a workflow profile, reducing per-spawn ski
 
 ## Implementation rules
 
-> **MANDATORY — .claude/ path bypass (R010)**
-> ALL writes to `.claude/` paths MUST go through a `/tmp/*.sh` script via Bash.
-> Direct `Write(".claude/...")` or `Edit(".claude/...")` triggers user approval prompts
-> that block automation regardless of `bypassPermissions` mode.
->
-> Pattern:
-> 1. Write tool → `/tmp/profile-{op}-$$.sh` with heredoc targeting `.claude/`
-> 2. Bash tool → `bash /tmp/profile-{op}-$$.sh`
->
-> This applies to `.claude/profiles/.active` and `~/.claude/settings.json` writes.
-> Reference: R010 (MUST-orchestrator-coordination.md), #1046, #1052, #1062.
+> **.claude/ path handling (CC v2.1.121+)**
+> Direct Write/Edit on `.claude/profiles/.active` and `~/.claude/settings.json` is permitted under `mode: "bypassPermissions"` (CC v2.1.121+, #1101). The legacy `/tmp/*.sh` bypass is deprecated. For CC < v2.1.121, see git history for the legacy pattern.
 
 ## Profiles directory
 

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

@@ -35,6 +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 |
 | Action (implement/generate) | sonnet | Optimized for code generation, balanced cost |
 | Post-verification (review/test) | sonnet or haiku | Structural verification, checklist validation |
 
@@ -62,3 +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 |
+
+## Opus 4.7 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:
+
+| Aspect | Opus 4.6 | Opus 4.7 |
+|--------|----------|----------|
+| 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
+
+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
+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)

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

@@ -58,13 +58,13 @@ Before execution, show the plan:
 
 ```
 [Scout] {url}
-├── Phase 1: Fetch & Summarize
-├── Phase 2: Load Project Philosophy
-├── Phase 3: Fit Analysis (sonnet)
-└── Phase 4: Issue Creation
+├── Phase 1: 콘텐츠 수집 및 요약
+├── Phase 2: 프로젝트 철학 로드
+├── Phase 3: 적합성 분석 (sonnet)
+└── Phase 4: 이슈 생성
 
-Estimated: ~1 min | Cost: ~$0.5-1.5
-Execute? [Y/n]
+예상: ~1분 | 비용: ~$0.5-1.5
+실행하시겠습니까? [Y/n]
 ```
 
 ## Workflow
@@ -90,7 +90,7 @@ Execute? [Y/n]
 
 ### Phase 3: Fit Analysis
 
-Spawn 1 sonnet agent with the following analysis prompt.
+Spawn 1 sonnet agent with `mode: "bypassPermissions"` and the following analysis prompt.
 
 **Inputs**:
 - Fetched content summary (Phase 1)
@@ -133,6 +133,7 @@ Return a structured verdict:
 - philosophy_table: criterion/fit/rationale for each dimension
 - recommendation: specific integration plan or skip reason
 - next_steps: 2-3 actionable items
+- IMPORTANT: Write all analysis output in Korean. Technical terms, code references, label names, and skill/agent names stay in English.
 - escalation: true/false (INTERNALIZE + M/L effort = true)
 ```
 
@@ -163,32 +164,32 @@ gh issue close {number} -c "Auto-closed: scout verdict is SKIP"
 ### Issue Body Template
 
 ```markdown
-## Scout Report: {title}
+## Scout 리포트: {title}
 
-**Source**: {url}
-**Verdict**: {INTERNALIZE / INTEGRATE / SKIP}
-**Priority**: {P1 / P2 / P3}
+**출처**: {url}
+**판정**: {INTERNALIZE / INTEGRATE / SKIP}
+**우선순위**: {P1 / P2 / P3}
 
-## Summary
-{2-3 sentence summary of the external content}
+## 요약
+{외부 콘텐츠에 대한 2-3문장 요약}
 
-## Philosophy Alignment
-| Criterion | Fit | Rationale |
-|-----------|-----|-----------|
-| Compilation metaphor | {check/cross} | {explanation} |
-| Separation of concerns (R006) | {check/cross} | {explanation} |
-| Dynamic agent creation | {check/cross} | {explanation} |
-| Existing skill overlap | {check/cross} | {overlapping skills list} |
+## 프로젝트 철학 정합성
+| 기준 | 적합 | 근거 |
+|------|------|------|
+| Compilation metaphor | {check/cross} | {설명} |
+| Separation of concerns (R006) | {check/cross} | {설명} |
+| Dynamic agent creation | {check/cross} | {설명} |
+| 기존 스킬 중복 | {check/cross} | {중복 스킬 목록} |
 
-## Recommendation
-{Specific integration plan — which skill/agent/guide to create, or why to skip}
+## 권장 사항
+{구체적 통합 계획 — 어떤 skill/agent/guide를 생성할지, 또는 건너뛰는 이유}
 
-## Next Steps
-- [ ] {follow-up action 1}
-- [ ] {follow-up action 2}
+## 다음 단계
+- [ ] {후속 조치 1}
+- [ ] {후속 조치 2}
 
 ---
-Generated by `/scout`
+`/scout`에 의해 생성됨
 ```
 
 ## Escalation
@@ -196,18 +197,18 @@ Generated by `/scout`
 When verdict is `INTERNALIZE` and integration effort is M or L:
 
 ```
-[Advisory] Deep analysis recommended.
-└── Consider running: /research {url}
+[Advisory] 심층 분석 권장.
+└── 실행 검토: /research {url}
 ```
 
 ## Result Display
 
 ```
-[Scout Complete] {title}
-├── Verdict: {INTERNALIZE / INTEGRATE / SKIP}
-├── Priority: {P1 / P2 / P3}
-├── Issue: #{number}
-└── Escalation: {/research recommended | none}
+[Scout 완료] {title}
+├── 판정: {INTERNALIZE / INTEGRATE / SKIP}
+├── 우선순위: {P1 / P2 / P3}
+├── 이슈: #{number}
+└── 에스컬레이션: {/research 권장 | 없음}
 ```
 
 ## Model Selection

package/templates/.claude/skills/skills-sh-search/SKILL.md

@@ -43,15 +43,16 @@ Search the [skills.sh](https://skills.sh/) marketplace for reusable AI agent ski
    ├── Verify installation in .claude/skills/
    └── Check installed SKILL.md frontmatter
 
-4. Post-install adaptation
-   ├── Review installed SKILL.md frontmatter
-   ├── Add oh-my-customcode fields if missing:
-   │   ├── user-invocable: true|false
-   │   ├── model-invocable (if not present)
-   │   └── argument-hint (if applicable)
-   └── Add source metadata:
-       ├── source-type: skills-sh
-       └── source-origin: <owner/repo>
+4. Post-install adaptation (delegate to mgr-creator per R010 Protected Paths)
+   ├── Delegate SKILL.md frontmatter adaptation to mgr-creator:
+   │   ├── Add oh-my-customcode fields if missing:
+   │   │   ├── user-invocable: true|false
+   │   │   ├── model-invocable (if not present)
+   │   │   └── argument-hint (if applicable)
+   │   └── Add source metadata:
+   │       ├── source-type: skills-sh
+   │       └── source-origin: <owner/repo>
+   └── mgr-creator validates R006 frontmatter compliance
 
 5. Ontology sync
    ├── Notify: run "omcustom ontology build" to register new skill

package/templates/.claude/skills/update-external/SKILL.md

@@ -62,6 +62,13 @@ python:
 ## Workflow
 
 ```
+0. Self-update oh-my-customcode package
+   ├── Check npm registry for latest version
+   ├── Compare with current installed version
+   ├── Run "npm install -g oh-my-customcode@<latest>" if outdated
+   ├── Skip automatically for npx invocations (npx always fetches latest)
+   └── Failure is non-blocking — continue to Step 1 regardless
+
 1. Identify external resources
    ├── Scan index.yaml files
    ├── Find source.type = "external"

package/templates/.github/workflows/cc-release-monitor.yml

@@ -0,0 +1,182 @@
+name: cc-release-monitor
+
+on:
+  schedule:
+    - cron: "17 7 * * *"
+  workflow_dispatch:
+
+permissions:
+  issues: write
+  contents: read
+
+env:
+  MIN_PATCH: "151"
+
+jobs:
+  monitor:
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Fetch Claude Code releases and create issues
+        env:
+          GH_TOKEN: ${{ github.token }}
+          GITHUB_REPOSITORY: ${{ github.repository }}
+          MIN_PATCH: ${{ env.MIN_PATCH }}
+        run: |
+          python3 << 'PYEOF'
+          import json
+          import os
+          import re
+          import subprocess
+          import sys
+          import textwrap
+
+          GH_TOKEN = os.environ["GH_TOKEN"]
+          REPO = os.environ["GITHUB_REPOSITORY"]
+          MIN_PATCH = int(os.environ.get("MIN_PATCH", "151"))
+
+          print(f"=== cc-release-monitor: fetching Claude Code releases ===")
+          print(f"MIN_PATCH={MIN_PATCH}")
+
+          # ── Step 1: fetch all releases from anthropics/claude-code ──────────
+          result = subprocess.run(
+              ["gh", "api", "repos/anthropics/claude-code/releases", "--paginate"],
+              capture_output=True,
+              text=True,
+              env=os.environ,
+          )
+          if result.returncode != 0:
+              print(f"ERROR: gh api failed: {result.stderr.strip()}", file=sys.stderr)
+              sys.exit(1)
+
+          try:
+              releases = json.loads(result.stdout)
+          except json.JSONDecodeError as e:
+              print(f"ERROR: Failed to parse releases JSON: {e}", file=sys.stderr)
+              sys.exit(1)
+
+          # ── Step 2: filter — v2.1.NNN, no nightly, patch >= MIN_PATCH ───────
+          pattern = re.compile(r"^v2\.1\.(\d+)$")
+          candidates = []
+          for r in releases:
+              tag = r.get("tag_name", "")
+              if "nightly" in tag.lower():
+                  continue
+              m = pattern.match(tag)
+              if not m:
+                  continue
+              if int(m.group(1)) < MIN_PATCH:
+                  continue
+              candidates.append({
+                  "tag": tag,
+                  "published_at": r.get("published_at", ""),
+                  "html_url": r.get("html_url", ""),
+                  "body": (r.get("body") or "").strip(),
+              })
+
+          print(f"Found {len(candidates)} candidate release(s) matching filter.")
+
+          if not candidates:
+              print("Nothing to do.")
+              sys.exit(0)
+
+          # ── Step 3: dedup and create ─────────────────────────────────────────
+          created = 0
+          skipped = 0
+
+          for release in candidates:
+              version = release["tag"]          # e.g. "v2.1.151"
+              published_at = release["published_at"]
+              html_url = release["html_url"]
+              body_raw = release["body"]
+
+              issue_title = f"Claude Code {version}"
+
+              # Dedup: exact title match among all issues (open + closed)
+              check = subprocess.run(
+                  [
+                      "gh", "issue", "list",
+                      "--repo", REPO,
+                      "--state", "all",
+                      "--search", issue_title,
+                      "--json", "title",
+                  ],
+                  capture_output=True,
+                  text=True,
+                  env=os.environ,
+              )
+              if check.returncode != 0:
+                  print(
+                      f"WARNING: issue list failed for {version}: {check.stderr.strip()}",
+                      file=sys.stderr,
+                  )
+                  skipped += 1
+                  continue
+
+              existing = json.loads(check.stdout or "[]")
+              if any(i["title"] == issue_title for i in existing):
+                  print(f"SKIP  {version} — issue already exists")
+                  skipped += 1
+                  continue
+
+              # Truncate release body
+              MAX_BODY = 2000
+              if not body_raw:
+                  release_summary = "_No release notes provided._"
+              elif len(body_raw) > MAX_BODY:
+                  release_summary = body_raw[:MAX_BODY] + "... (truncated)"
+              else:
+                  release_summary = body_raw
+
+              issue_body = textwrap.dedent(f"""\
+                  # Claude Code {version}
+
+                  **Release:** {version}
+                  **Published:** {published_at}
+                  **Link:** {html_url}
+
+                  ## Release Summary
+
+                  {release_summary}
+
+                  ---
+
+                  ## Action Items
+
+                  - [ ] Review release notes for impact on oh-my-customcode
+                  - [ ] Update agent definitions if new Claude Code features affect agents
+                  - [ ] Test compatibility with current oh-my-customcode version
+                  - [ ] Update CLAUDE.md if new capabilities are relevant
+
+                  ---
+
+                  _This issue was auto-created by the cc-release-monitor GitHub Actions workflow._\
+              """)
+
+              create = subprocess.run(
+                  [
+                      "gh", "issue", "create",
+                      "--repo", REPO,
+                      "--title", issue_title,
+                      "--body", issue_body,
+                      "--label", "automated,claude-code-release",
+                  ],
+                  capture_output=True,
+                  text=True,
+                  env=os.environ,
+              )
+              if create.returncode != 0:
+                  print(
+                      f"ERROR  {version} — create failed: {create.stderr.strip()}",
+                      file=sys.stderr,
+                  )
+                  skipped += 1
+                  continue
+
+              issue_url = create.stdout.strip()
+              print(f"CREATE {version} — {issue_url}")
+              created += 1
+
+          print()
+          print(f"=== Summary: {created} created, {skipped} skipped ===")
+          PYEOF

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

@@ -1,7 +1,7 @@
 # Claude Code Version Compatibility
 
-> Updated: 2026-05-15
-> Source: Claude Code release notes (#967, #968, #969, #1126 auto-detected by claude-native skill, #1137, #1158)
+> Updated: 2026-05-29
+> Source: Claude Code release notes (#967, #968, #969, #1126 auto-detected by claude-native skill, #1137, #1158, #1242, #1243, #1244, #1245)
 
 ## Compatibility Baseline
 
@@ -491,6 +491,11 @@ docs/superpowers/plans/**
 | v2.1.148 | Bash 도구 exit 127 regression(v2.1.147 도입) 수정 — v2.1.147 사용 시 즉시 업그레이드 권장. | None |
 | v2.1.149 | `/usage` per-category(skills/subagents/plugins/MCP) breakdown, GFM 체크박스 렌더링, worktree sandbox allowlist fix, `find` macOS vnode crash fix. | P3 follow-up |
 | v2.1.150 | 내부 인프라 개선만, 사용자 대면 변경 없음. 조치 불필요. | None |
+| v2.1.152 | `disallowed-tools` in skill frontmatter (R002/R006), `/reload-skills`, `SessionStart reloadSkills` — no harness change needed. | None |
+| v2.1.153 | statusline `COLUMNS`/`LINES` env (R012), `skipLfs` marketplace option, `claude agents` autocomplete improvement. | P3 follow-up |
+| v2.1.154 | Opus 4.8 + `opus48` alias (R006), dynamic workflows (R009/R018), lean system prompt default, `CLAUDE_CODE_OPUS_4_6_FAST_MODE_OVERRIDE` deprecated 06/01. | P3 follow-up |
+| v2.1.156 | Opus 4.8 thinking-block API-error fix — no harness change needed. | None |
+| #1241 | Agent tool malformed-parsing on long/special-char prompts — mitigations documented in Known Platform Issues. | See below |
 
 ## v2.1.144 (2026-05-19) — 호환성 점검
 
@@ -889,6 +894,233 @@ Bash 도구의 `find` 명령이 macOS system file/vnode table을 소진하여
 
 ---
 
+## v2.1.152 (2026-05-27) — 호환성 점검
+
+> Issue: #1242 — CC v2.1.152 tracking
+
+### Skill/slash-command `disallowed-tools` frontmatter (R002/R006)
+
+Skills and slash commands can now declare `disallowed-tools` in their frontmatter to remove specific tools from the model while the skill is active.
+
+**oh-my-customcode 연관**: R002 tool permission tiers 및 R006 skill frontmatter optional fields와 직접 연관됩니다. 특정 스킬 실행 구간에 Bash나 Write 등 위험 도구를 제거하는 세밀한 tool scope 제어가 가능해집니다. 예: 조사 전용 스킬에서 `disallowed-tools: [Bash, Write]` 선언으로 read-only 강제. 신규 스킬 작성 시 활용 권장.
+
+### `/reload-skills` 명령 + `SessionStart reloadSkills: true`
+
+`/reload-skills` 커맨드로 세션 재시작 없이 스킬 디렉토리를 재스캔할 수 있습니다. `SessionStart` 훅이 `reloadSkills: true`를 반환하면 훅이 설치한 스킬이 동일 세션 내에서 즉시 사용 가능합니다.
+
+**oh-my-customcode 연관**: R006 스킬 인프라 및 SessionStart 훅(`hooks.json`)과 연관됩니다. 현재 SessionStart 훅은 HUD 초기화 및 메모리 로드에 사용되는데, 훅에서 스킬을 동적 설치하는 패턴이 필요한 경우 `reloadSkills: true`로 즉시 적용 가능합니다. 현재 하네스에서 즉시 활용 계획 없음 — 직접 변경 불필요.
+
+### `SessionStart` 훅 세션 제목 설정
+
+`SessionStart` 훅이 `hookSpecificOutput.sessionTitle`을 통해 세션 제목을 설정할 수 있습니다.
+
+**oh-my-customcode 연관**: R012 HUD와 결합하여 프로젝트/에이전트 컨텍스트를 창 제목에 반영하는 것이 가능합니다. 선택적 활용.
+
+### `MessageDisplay` 훅 이벤트
+
+어시스턴트 메시지 텍스트를 표시 시점에 변환하거나 숨길 수 있는 새 훅 이벤트입니다.
+
+**oh-my-customcode 연관**: 고급 HUD 커스터마이제이션 후보. `.claude/hooks/` 변경은 R001에 따라 사용자 명시 승인 필요.
+
+### 기타 변경
+
+- `--fallback-model` 설정 시 주 모델 불가 상태에서 세션 내 자동 fallback (오류 없이 계속)
+- Auto mode opt-in consent 제거 (자동화 환경 진입 간소화)
+- 터미널 렌더링 안정성 개선 다수 (툴 결과 링크 클릭, 마크다운 테이블, 포커스 모드 등)
+
+**oh-my-customcode 연관**: `--fallback-model` 패턴은 R006 `model:` 프론트매터의 보완 수단으로 CI 환경에서 활용 가능합니다. 직접 harness 변경 불필요.
+
+### oh-my-customcode 연관 평가
+
+| 변경 | oh-my-customcode 영향 | 조치 |
+|------|----------------------|------|
+| Skill `disallowed-tools` frontmatter | R002/R006 tool scope 세밀 제어 | 신규 스킬에서 선택적 활용 (opt-in) |
+| `/reload-skills` + `SessionStart reloadSkills` | R006 스킬 동적 로딩 | None (현재 즉각 활용 계획 없음) |
+| `SessionStart` 세션 제목 설정 | R012 HUD 강화 후보 | P3 follow-up |
+| `MessageDisplay` 훅 | HUD 고급 커스터마이제이션 | P3 follow-up |
+| `--fallback-model` | CI 환경 모델 안정성 | None (opt-in) |
+| 터미널 렌더링 수정 | UX 개선 | None |
+
+**Action items**:
+- 본 릴리스에서 코드 변경 없음 (docs-only)
+- 후속 follow-up 후보:
+  1. `disallowed-tools` frontmatter를 read-only 스킬(연구·분석)에 적용 검토
+  2. `SessionStart sessionTitle` hook을 R012 HUD 창 제목 통합에 활용 검토
+
+---
+
+## v2.1.153 (2026-05-28) — 호환성 점검
+
+> Issue: #1243 — CC v2.1.153 tracking
+
+### Statusline scripts `COLUMNS`/`LINES` 환경 변수 수신 (R012)
+
+`.claude/statusline.sh`를 포함한 statusline 스크립트가 `COLUMNS`와 `LINES` 환경 변수를 받게 되어 출력을 터미널 너비에 맞게 조정할 수 있습니다.
+
+**oh-my-customcode 연관**: R012 HUD Statusline과 직접 연관됩니다. 현재 `.claude/statusline.sh`는 고정 형식으로 출력되는데, 좁은 터미널에서 상태줄이 잘리는 문제를 `COLUMNS`를 활용하여 반응형으로 개선할 수 있습니다. 별도 follow-up 권장 (P3).
+
+### `skipLfs` 플러그인 마켓플레이스 옵션
+
+`github`/`git` 플러그인 마켓플레이스 소스에 `skipLfs: true` 옵션이 추가되어 Git LFS 다운로드를 건너뛸 수 있습니다.
+
+**oh-my-customcode 연관**: 대형 LFS 에셋을 포함한 플러그인을 사용하는 환경에서 설치 속도 개선에 활용 가능합니다. 현재 oh-my-customcode 마켓플레이스 설정에는 직접 영향 없음.
+
+### `claude agents` 자동완성 개선
+
+dispatch 입력의 자동완성이 프로젝트 스킬뿐 아니라 네이티브 슬래시 커맨드와 번들 스킬도 제안합니다.
+
+**oh-my-customcode 연관**: R019 라우팅 및 에이전트 디스패치 UX 개선. 네이티브 `/code-review`와 내부 `dev-review` 스킬 구분이 자동완성에서도 명확해집니다. 직접 harness 변경 불필요.
+
+### `/model` 기본값 저장 (`s` 키로 현재 세션만 변경)
+
+`/model` 선택이 새 세션 기본값으로 저장됩니다. 현재 세션만 변경하려면 `s` 키를 사용합니다. 기존 `d` 키는 `s` 키(`thisSessionOnly`)로 변경되었습니다.
+
+**oh-my-customcode 연관**: R006 agent frontmatter `model:` 설정이 세션 기본값과 독립적으로 동작하는 기존 동작과 정합. keybindings.json에서 `modelPicker:setAsDefault`를 커스터마이즈한 경우 `modelPicker:thisSessionOnly`로 rename 필요.
+
+### 기타 수정 (안정성)
+
+- MCP 서버 frontmatter policy 적용 수정 (`--strict-mcp-config`, managed settings 정책)
+- 과도한 메모리 사용 수정 (세션 파일 경로로 세션 재개 시 여러 GB)
+- `Agent` 도구 `subagent_type: 'claude'` 사용 시 gitignored 경로 출력 무음 폐기 수정
+- stateful MCP 서버 reconnect-loop regression 수정 (v2.1.147 도입)
+- 백그라운드 세션 다수 UX/안정성 수정
+
+**oh-my-customcode 연관**: MCP 정책 수정은 R011 claude-mem / R019 ontology-rag 서버 동작에 영향 가능합니다. `subagent_type: 'claude'` gitignored 경로 수정은 R010 subagent 델리게이션 시 `.claude/outputs/` 아티팩트 채널 사용 신뢰성을 높입니다.
+
+### oh-my-customcode 연관 평가
+
+| 변경 | oh-my-customcode 영향 | 조치 |
+|------|----------------------|------|
+| Statusline `COLUMNS`/`LINES` | R012 반응형 statusline 가능 | P3 follow-up |
+| `skipLfs` 마켓플레이스 옵션 | LFS 포함 플러그인 설치 개선 | None |
+| `claude agents` 자동완성 | 네이티브/내부 스킬 구분 UX | None |
+| `/model` 기본값 저장 (`s` 키) | R006 model 설정 정합 | keybindings.json 커스터마이즈 시 rename |
+| MCP 서버 frontmatter policy 수정 | R011/R019 MCP 서버 정책 안정성 | 호환, 개선 |
+| MCP 과다 메모리 사용 수정 | 장시간 세션 안정성 | 호환, 개선 |
+| `subagent_type: 'claude'` gitignored 수정 | R010 아티팩트 채널 신뢰성 | 호환, 개선 |
+
+**Action items**:
+- 본 릴리스에서 코드 변경 없음 (docs-only)
+- 후속 follow-up 후보:
+  1. `.claude/statusline.sh`에 `COLUMNS` 기반 반응형 출력 적용 (R012)
+
+---
+
+## v2.1.154 (2026-05-28) — 호환성 점검
+
+> Issue: #1244 — CC v2.1.154 tracking
+
+> **Note**: v2.1.151 및 v2.1.155는 공개 릴리즈가 없었습니다 (Claude Code는 일부 패치 번호를 건너뜁니다).
+
+### Opus 4.8 도입 (R006 model aliases)
+
+Opus 4.8이 출시되었습니다. 기본 effort는 `high`이며, `/effort xhigh`로 최고 추론 강도를 활성화할 수 있습니다. Fast Mode on Opus 4.8이 표준 요금의 2배 가격에 2.5배 속도로 제공됩니다.
+
+**oh-my-customcode 연관**: R006 model aliases 표에 `opus48` alias 추가가 필요합니다. 현재 `opus` alias는 `claude-opus-4-6`을 가리키며, Opus 4.8 사용 에이전트는 full model ID(`claude-opus-4-8`) 또는 신규 alias를 명시해야 합니다. `CLAUDE_CODE_OPUS_4_6_FAST_MODE_OVERRIDE` 환경 변수는 **2026-06-01 폐기 예정** — 해당 변수를 사용하는 환경에서는 `/model claude-opus-4-6[1m]` + `/fast on`으로 전환 필요합니다.
+
+**추천 업데이트**: R006 MUST-agent-design.md의 model aliases 표에 `opus48` → `claude-opus-4-8` 항목 추가. 별도 이슈로 추적 권장.
+
+### Dynamic Workflows — `Workflow` 도구 (R009/R018)
+
+`Workflow` 도구가 일반 공개되었습니다. `/workflows`로 실행 중인 워크플로우를 확인할 수 있습니다. 백그라운드에서 수십~수백 개의 에이전트를 오케스트레이션합니다.
+
+**oh-my-customcode 연관**: R009 병렬 실행 및 R018 Agent Teams와 개념적으로 중첩됩니다. oh-my-customcode `/pipeline` 스킬(workflows/*.yaml)과의 통합 가능성이 높아졌습니다. 현재는 네이티브 `Workflow` 도구를 실험적으로 평가하는 단계이며, R009/R018 기존 패턴이 기본으로 유지됩니다. 중장기 통합 후보 (P2).
+
+### Lean system prompt 기본값 변경
+
+Lean system prompt가 Haiku, Sonnet, Opus 4.7 이하를 제외한 모든 모델의 기본값이 되었습니다.
+
+**oh-my-customcode 연관**: Opus 4.8 기반 에이전트(arch-documenter, arch-speckit-agent 등)는 lean system prompt로 실행됩니다. system prompt 토큰이 감소하여 R013 ecomode context budget에 긍정적입니다. 직접 harness 변경 불필요.
+
+### `AskUserQuestion` 자제 개선
+
+Claude가 스스로 판단할 수 있는 상황에서는 multiple-choice prompt를 보내지 않도록 개선되었습니다.
+
+**oh-my-customcode 연관**: R015 intent transparency의 ambiguity-gate 패턴과 정합. 불필요한 확인 요청 감소로 자동화 흐름이 개선됩니다.
+
+### `/simplify` 역할 변경
+
+`/simplify`가 이제 cleanup-only review(재사용, 단순화, 효율성)를 실행하고 수정사항을 적용합니다. `/code-review --fix`의 전체 버그 탐색 리뷰와는 별도입니다.
+
+**oh-my-customcode 연관**: v2.1.146/v2.1.147에서 언급된 `/simplify`→`/code-review` 개명의 후속 조치입니다. 내부 `dev-review` 스킬과는 여전히 충돌 없음.
+
+### `claude agents`: shell 명령 백그라운드 세션으로 실행
+
+`claude agents`에서 `! <command>`를 입력하면 셸 명령이 독립 백그라운드 세션으로 실행됩니다. `claude --bg --exec '<command>'`와 동일합니다.
+
+**oh-my-customcode 연관**: 긴 CI/빌드 명령을 백그라운드로 분리할 때 유용합니다. R010 Universal bypassPermissions와 무관 (별도 채널).
+
+### 기타 변경
+
+- Plugins `defaultEnabled: false` — 필요 시 명시 활성화 가능 (oh-my-customcode 필수 플러그인 영향 없음)
+- Streaming tool execution 항상 활성화 (Bedrock/Vertex/Foundry 포함)
+- `Stdio MCP` 서브프로세스에 `CLAUDE_CODE_SESSION_ID`, `CLAUDECODE=1` 환경 변수 전달
+- `rm -rf $HOME` trailing slash 경우 위험 경로 차단 수정 (R001 보안 강화)
+- 배경 에이전트 worktree 격리 guard 수정 (공유 체크아웃 write 방지)
+
+**oh-my-customcode 연관**: `CLAUDE_CODE_SESSION_ID` 전달은 R011 메모리 MCP 서버(`claude-mem`, `agentmemory`)의 세션 스코핑에 활용 가능합니다. `rm -rf $HOME` trailing slash 수정은 R001 안전 규칙 강화입니다.
+
+### oh-my-customcode 연관 평가
+
+| 변경 | oh-my-customcode 영향 | 조치 |
+|------|----------------------|------|
+| Opus 4.8 도입 | R006 model aliases 업데이트 필요 | 별도 이슈 등록 권장 (P2) |
+| `CLAUDE_CODE_OPUS_4_6_FAST_MODE_OVERRIDE` 폐기 (06/01) | 해당 변수 사용 환경 전환 필요 | 06/01 이전 조치 필요 |
+| Dynamic Workflows | R009/R018 + `/pipeline` 스킬 통합 후보 | P2 follow-up |
+| Lean system prompt 기본값 | Opus 4.8 에이전트 토큰 감소 | None (자동 적용) |
+| `AskUserQuestion` 자제 | R015 ambiguity-gate 개선 | None |
+| `/simplify` cleanup-only | 내부 `dev-review`와 충돌 없음 | None |
+| `! <command>` bg shell | 빌드 명령 백그라운드 분리 | None (opt-in) |
+| `CLAUDE_CODE_SESSION_ID` MCP 전달 | R011 MCP 세션 스코핑 | None (옵션 활용) |
+| `rm -rf $HOME` trailing slash fix | R001 보안 강화 | None |
+| 배경 에이전트 worktree guard fix | R009 격리 안전성 | None |
+
+**Action items**:
+- **[P2]** R006 MUST-agent-design.md model aliases 표에 `opus48` → `claude-opus-4-8` 추가 (별도 이슈 등록 권장)
+- **[긴급]** `CLAUDE_CODE_OPUS_4_6_FAST_MODE_OVERRIDE` 사용 환경이 있다면 2026-06-01 전에 `/model claude-opus-4-6[1m]` + `/fast on` 패턴으로 전환
+- **[P2]** Dynamic Workflows와 `/pipeline` 스킬 통합 가능성 연구
+
+---
+
+## v2.1.156 (2026-05-29) — 호환성 점검
+
+> Issue: #1245 — CC v2.1.156 tracking
+
+> **Note**: v2.1.155는 공개 릴리즈가 없었습니다 (Claude Code는 일부 패치 번호를 건너뜁니다).
+
+### Opus 4.8 thinking-block API 오류 수정
+
+Opus 4.8에서 thinking blocks가 수정되어 API 오류가 발생하던 버그가 수정되었습니다.
+
+**oh-my-customcode 연관**: Opus 4.8 기반 에이전트(v2.1.154에서 도입)의 안정성이 개선됩니다. v2.1.154로 Opus 4.8을 도입한 경우 v2.1.156으로 업그레이드가 강력히 권장됩니다. 직접 harness 변경 불필요.
+
+### oh-my-customcode 연관 평가
+
+| 변경 | oh-my-customcode 영향 | 조치 |
+|------|----------------------|------|
+| Opus 4.8 thinking-block API 오류 수정 | Opus 4.8 에이전트 안정성 복구 | v2.1.154 사용 시 즉시 업그레이드 |
+
+**Action items**:
+- v2.1.154를 사용 중이라면 v2.1.156으로 즉시 업그레이드 (Opus 4.8 thinking-block 안정성)
+- 본 릴리스에서 코드 변경 없음 (docs-only)
+
+---
+
+## Known Platform Issues & Workarounds
+
+### Agent tool malformed parsing on long / special-character prompts (#1241)
+
+**Symptom**: The Agent (subagent dispatch) tool intermittently fails to parse delegation prompts when they are very long or contain heavy special characters (backticks, consecutive colons, shell-variable syntax), reporting the tool call as `malformed`. This is a Claude Code platform-level serialization issue, not an oh-my-customcode defect.
+
+**Workarounds** (already aligned with existing rules):
+- Pre-decompose oversized delegations: when a delegated prompt exceeds ~5000 tokens or spans 3+ unrelated domains, split into parallel domain-scoped agents (R009 giant-prompt anti-pattern).
+- In delegation prompts, prefer plain prose over heavy literal blocks; avoid long fenced code with backticks and consecutive `::` sequences where a plain description suffices.
+- For unavoidable large payloads, hand off via an artifact file path (R006 Artifact Channel Protocol) instead of inlining the content in the prompt.
+
+> Issue: #1241 — Agent tool malformed-parsing workaround (platform bug; mitigations documented)
+
+---
+
 ## References
 
 - #967 — Claude Code v2.1.117 release note
@@ -906,6 +1138,11 @@ Bash 도구의 `find` 명령이 macOS system file/vnode table을 소진하여
 - #1218 — Claude Code v2.1.148 compatibility documentation
 - #1219 — Claude Code v2.1.149 compatibility documentation
 - #1220 — Claude Code v2.1.150 compatibility documentation
+- #1241 — Agent tool malformed-parsing workaround
+- #1242 — Claude Code v2.1.152 compatibility documentation
+- #1243 — Claude Code v2.1.153 compatibility documentation
+- #1244 — Claude Code v2.1.154 compatibility documentation
+- #1245 — Claude Code v2.1.156 compatibility documentation
 - `.claude/skills/claude-native/` — auto-generation source
 - `.claude/rules/SHOULD-hud-statusline.md` — R012 statusline integration
 - `.claude/rules/MUST-agent-design.md` — R006 agent frontmatter spec

package/templates/manifest.json

@@ -1,5 +1,5 @@
 {
-  "version": "0.154.1",
+  "version": "0.156.0",
   "lastUpdated": "2026-05-20T00:00:00.000Z",
   "omcustomMinClaudeCode": "2.1.121",
   "omcustomMinClaudeCodeReason": "Sensitive-path direct Write/Edit on .claude/** under bypassPermissions (R010 deprecation, #1101)",