oh-my-customcode 0.158.0 → 0.160.0

package/README.md

@@ -13,7 +13,7 @@
 
 **[한국어 문서 (Korean)](./README_ko.md)**
 
-49 agents. 118 skills. 23 rules. One command.
+49 agents. 115 skills. 23 rules. One command.
 
 ```bash
 npm install -g oh-my-customcode && cd your-project && omcustom init
@@ -132,7 +132,7 @@ Each agent declares its tools, model, memory scope, and limitations in YAML fron
 
 ---
 
-### Skills (118)
+### Skills (115)
 
 | Category | Count | Includes |
 |----------|-------|----------|
@@ -145,7 +145,7 @@ Each agent declares its tools, model, memory scope, and limitations in YAML fron
 | Package | 3 | npm-publish, npm-version, npm-audit |
 | Optimization | 3 | optimize-analyze, optimize-bundle, optimize-report |
 | Security | 3 | adversarial-review, cve-triage, jinja2-prompts |
-| Other | 10 | codex-exec, claude-native, vercel-deploy, skills-sh-search, result-aggregation, writing-clearly-and-concisely, and more |
+| Other | 7 | claude-native, vercel-deploy, skills-sh-search, result-aggregation, writing-clearly-and-concisely, and more |
 
 Skills use a 3-tier scope system: `core` (universal), `harness` (agent/skill maintenance), `package` (project-specific).
 
@@ -270,7 +270,7 @@ your-project/
 ├── CLAUDE.md                   # Entry point
 ├── .claude/
 │   ├── agents/                 # 49 agent definitions
-│   ├── skills/                 # 118 skill modules
+│   ├── skills/                 # 115 skill modules
 │   ├── rules/                  # 22 governance rules (R000-R021)
 │   ├── hooks/                  # 15 lifecycle hook scripts
 │   ├── schemas/                # Tool input validation schemas

package/dist/cli/index.js

@@ -2334,7 +2334,7 @@ var init_package = __esm(() => {
     workspaces: [
       "packages/*"
     ],
-    version: "0.158.0",
+    version: "0.160.0",
     description: "Batteries-included agent harness for Claude Code",
     type: "module",
     bin: {

package/dist/index.js

@@ -2031,7 +2031,7 @@ var package_default = {
   workspaces: [
     "packages/*"
   ],
-  version: "0.158.0",
+  version: "0.160.0",
   description: "Batteries-included agent harness for Claude Code",
   type: "module",
   bin: {

package/package.json

@@ -3,7 +3,7 @@
   "workspaces": [
     "packages/*"
   ],
-  "version": "0.158.0",
+  "version": "0.160.0",
   "description": "Batteries-included agent harness for Claude Code",
   "type": "module",
   "bin": {

package/templates/.claude/rules/MUST-agent-design.md

@@ -72,7 +72,7 @@ domain: backend              # backend | frontend | data-engineering | devops |
 disableSkillShellExecution: true  # Disable inline shell execution in skills (v2.1.91+)
 ```
 
-> **Note**: When `disableSkillShellExecution` is enabled (v2.1.91+), skills that rely on inline shell execution (e.g., `codex-exec`, `gemini-exec`, `rtk-exec`) will have their shell blocks disabled. This is a security hardening option.
+> **Note**: When `disableSkillShellExecution` is enabled (v2.1.91+), skills that rely on inline shell execution (e.g., `rtk-exec`) will have their shell blocks disabled. This is a security hardening option.
 -->
 
 <!-- DETAIL: CC Version Compatibility History

package/templates/.claude/rules/MUST-agent-teams.md

@@ -23,6 +23,7 @@ Available when `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1` or TeamCreate/SendMessag
 | Dynamic agent creation + usage | **Agent Teams** | Create → test → iterate cycle |
 | Multi-issue release batch | **Agent Teams** | Shared task tracking, coordinated release |
 | Large plan / multi-domain prompt (>5000 tokens, 3+ areas) | **Agent Teams** | Domain-split parallel writing + review loop avoids single-agent timeout |
+| Mechanical disjoint-file refactoring (bulk delete + reference cleanup) | Agent Tool | Pure parallel edits with no peer coordination or review loop; Teams member-stall risk outweighs benefit — use standalone parallel Agents (R009) |
 
 **When Agent Teams is enabled and criteria are met, usage is required.**
 
@@ -188,7 +189,7 @@ The skill's steps are followed, but agent spawning uses Teams when criteria are
 ## Team Patterns
 
 Standard: Research (researcher-1 + researcher-2 + synthesizer), Development (implementer + reviewer + tester), Debug (investigator-1 + investigator-2 + fixer).
-Hybrid: Review+Fix, Create+Validate, Multi-Expert, Dynamic Creation, Codex Hybrid.
+Hybrid: Review+Fix, Create+Validate, Multi-Expert, Dynamic Creation.
 
 <!-- DETAIL: Team Patterns
 ### Standard Patterns
@@ -203,21 +204,6 @@ Hybrid: Review+Fix, Create+Validate, Multi-Expert, Dynamic Creation, Codex Hybri
 
 ### Dynamic Patterns
 - Dynamic Creation: mgr-creator + domain-expert (create new agent → immediately use for pending task)
-- Codex Hybrid: codex-exec-agent + claude-reviewer (Codex generates → Claude reviews/refines)
-
-### Codex-Exec Integration
-When both Agent Teams and codex-exec are available:
-  1. Claude agent analyzes requirements
-  2. codex-exec generates implementation (Codex strength: code generation)
-  3. Claude agent reviews and refines (Claude strength: reasoning, quality)
-  4. Iterate via team messaging until quality meets standards
-
-| Step | Agent | Model |
-|------|-------|-------|
-| Analysis | Claude team member | sonnet/opus |
-| Generation | codex-exec | (Codex default) |
-| Review | Claude team member | sonnet |
-| Refinement | Appropriate expert | sonnet |
 
 ### Dynamic Agent Creation in Teams
 When Agent Teams creates a new agent via mgr-creator:
@@ -328,3 +314,27 @@ Agent Teams 멤버는 long-running 작업 중 진행 상태를 TaskUpdate 로
 - 차단 사유를 SendMessage 로만 보내고 task description 업데이트 누락 → TaskList 만 보는 멤버는 사유를 모름
 
 Reference issue: #1087.
+
+## Member Completion Verification (deterministic ground-truth)
+
+Agent Teams member completion MUST be verified by deterministic ground-truth — NOT by SendMessage reports or TaskList status alone. Members may edit files without updating task status (task stays `pending`) or go idle without executing at all.
+
+**Verification sources (in order of reliability):**
+
+| Source | Reliability | Examples |
+|--------|-------------|---------|
+| `git status` / `git diff` | High — ground truth | Check that expected files changed |
+| `grep` / file existence | High — deterministic | Verify expected content written |
+| Validation scripts | High — deterministic | `validate-docs`, linters, test runs |
+| TaskList status | Low — member may not update | Use as a signal only |
+| SendMessage report | Low — member may stall before sending | Use as a signal only |
+
+Cross-reference: R020 ("actual outcome ≠ attempt" — verifying that a command ran is not the same as verifying it succeeded).
+
+**Stall handling**: When a member shows no task progress within ~2 minutes despite spawn + owner assignment + SendMessage coordination, reassign the work to a standalone Agent (R009) rather than continuing to nudge the stalled member. Stalled Teams members waste tokens on idle polling and delay the overall workflow.
+
+Observed instance: v0.159.0 release (session 105) — members assigned to disjoint-file cleanup tasks went idle without executing; deterministic git-diff check exposed the gap; work was reassigned to standalone parallel Agents. References: #1261, #1262.
+
+## Member Prompt Size Cap
+
+Keep per-member delegation prompts under ~5000 tokens and within a single domain. Oversized or multi-domain prompts risk malformed-parsing truncation in the CC platform (see R009 giant-prompt heuristic and `feedback_agent_malformed_parsing.md`). Large multi-file delegations should be decomposed and split across multiple members or standalone Agents.

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

@@ -70,27 +70,7 @@ Check if Agent Teams is available (`CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1` or T
 | Cross-tool data quality analysis | Agent Teams |
 | Quick DAG/model validation | Agent Tool |
 
-### Step 2: Codex-Exec Hybrid (Code Generation)
-For **new pipeline code**, **DAG scaffolding**, or **SQL model generation**:
-
-1. Check `/tmp/.claude-env-status-*` for codex, gemini, and rtk availability
-2. If codex available AND task involves new file creation → automatically delegate to `/codex-exec` for scaffolding:
-   - Display: `[Codex Hybrid] Delegating to codex-exec...`
-   - codex-exec generates initial code (strength: fast generation)
-   - Selected DE expert reviews and refines codex output (strength: reasoning, quality)
-3. If codex unavailable but gemini available → delegate to `/gemini-exec` for scaffolding:
-   - Display: `[Gemini Hybrid] Delegating to gemini-exec...`
-   - gemini-exec generates initial code
-   - Selected DE expert reviews and refines output
-4. If RTK available (`RTK=available` in env status) → optionally wrap DE expert output through RTK to reduce token consumption by 60-90%:
-   - Display: `[RTK Proxy] Token optimization active via rtk-exec`
-   - RTK acts as a transparent proxy — no change to expert selection
-5. If none available → display `[External CLI] Unavailable — proceeding with {expert} directly` and use DE expert directly
-
-**Suitable**: New DAG files, dbt model scaffolding, SQL template generation
-**Unsuitable**: Existing pipeline modification, architecture decisions, data quality analysis
-
-### Step 3: Expert Selection
+### Step 2: Expert Selection
 Route to appropriate DE expert based on tool/framework detection.
 
 > **Permission Mode**: When spawning agents via Agent tool, always pass `mode: "bypassPermissions"`. The Agent tool default (`acceptEdits`) overrides agent frontmatter `permissionMode`, causing permission prompts during unattended execution.

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

@@ -109,27 +109,7 @@ Check if Agent Teams is available (`CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1` or T
 | Cross-layer debugging (FE + BE + DB) | Agent Teams |
 | Simple file search/validation | Task Tool |
 
-### Step 2: Codex-Exec Hybrid (Implementation Tasks)
-For **new file creation**, **boilerplate**, or **test code generation**:
-
-1. Check `/tmp/.claude-env-status-*` for codex, gemini, and rtk availability
-2. If codex available AND task involves new file creation → automatically delegate to `/codex-exec` for scaffolding:
-   - Display: `[Codex Hybrid] Delegating to codex-exec...`
-   - codex-exec generates initial code (strength: fast generation)
-   - Selected Claude expert reviews and refines codex output (strength: reasoning, quality)
-3. If codex unavailable but gemini available → delegate to `/gemini-exec` for scaffolding:
-   - Display: `[Gemini Hybrid] Delegating to gemini-exec...`
-   - gemini-exec generates initial code
-   - Selected Claude expert reviews and refines output
-4. If RTK available (`RTK=available` in env status) → optionally wrap Claude expert output through RTK to reduce token consumption by 60-90%:
-   - Display: `[RTK Proxy] Token optimization active via rtk-exec`
-   - RTK acts as a transparent proxy — no change to expert selection
-5. If none available → display `[External CLI] Unavailable — proceeding with {expert} directly` and use Claude expert directly
-
-**Suitable**: New file creation, boilerplate, scaffolding, test code
-**Unsuitable**: Existing code modification, architecture decisions, bug fixes
-
-### Step 3: Expert Agent Selection
+### Step 2: Expert Agent Selection
 Route to appropriate language/framework expert based on file extension and keyword mapping.
 
 > **Permission Mode**: When spawning agents via Agent tool, always pass `mode: "bypassPermissions"`. The Agent tool default (`acceptEdits`) overrides agent frontmatter `permissionMode`, causing permission prompts during unattended execution.

package/templates/.claude/skills/intent-detection/SKILL.md

@@ -277,14 +277,8 @@ english:
 ```
 Research intent detected (confidence >= 70%)
     ↓
-Check Codex CLI availability
-    ├─ Available (codex binary + OPENAI_API_KEY)
-    │   → Use codex-exec skill with --effort xhigh
-    │   → Prompt: "Research and analyze: {user_request}"
-    │   → Returns: structured findings for orchestrator
-    └─ Unavailable
-        → Fall back to Claude's WebFetch/WebSearch
-        → Orchestrator handles directly or via general-purpose agent
+Use Claude's WebFetch/WebSearch
+→ Orchestrator handles directly or via general-purpose agent
 ```
 
 ### Confidence Scoring
@@ -303,7 +297,7 @@ Check Codex CLI availability
 ├── Input: "{user input}"
 ├── Workflow: research-workflow
 ├── Confidence: {percentage}%
-├── Method: codex-exec (xhigh) | WebFetch fallback
+├── Method: WebFetch/WebSearch
 └── Reason: {explanation}
 ```
 

package/templates/.claude/skills/intent-detection/patterns/agent-triggers.yaml

@@ -338,10 +338,10 @@ agents:
       - gather
     base_confidence: 50
     action_weight: 30
-    routing_rule: "MUST use codex-exec --effort xhigh when codex available, or gemini-exec when gemini available, fallback to WebFetch/WebSearch"
+    routing_rule: "Use WebFetch/WebSearch for research; orchestrator handles directly or via general-purpose agent"
 
   # ---------------------------------------------------------------------------
-  # Code Generation (hybrid workflow, skill-based)
+  # Code Generation (skill-based)
   # ---------------------------------------------------------------------------
   code-generation-workflow:
     keywords:
@@ -367,7 +367,7 @@ agents:
       - scaffold
     base_confidence: 30
     action_weight: 25
-    routing_rule: "Suggest codex-exec hybrid when codex available, or gemini-exec hybrid when gemini available, for new file creation tasks"
+    routing_rule: "Use Claude expert agents for new file creation tasks"
 
   # Managers (continued)
   mgr-gitnerd:
@@ -630,15 +630,6 @@ agents:
     base_confidence: 75
     routing_rule: "Invoke worker-reviewer-pipeline skill"
 
-  skill-agora:
-    keywords:
-      korean: [아고라, 합의, 적대적 합의, 다중 LLM]
-      english: [agora, consensus, "adversarial consensus", "multi llm"]
-    file_patterns: []
-    supported_actions: [debate, consensus, verify]
-    base_confidence: 85
-    routing_rule: "Invoke agora skill"
-
   # Debugging skill
   skill-debugging:
     keywords:

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

@@ -163,32 +163,17 @@ Batch 3: T9, T10            (Innovation)
 
 ### Phase 2: Cross-Verification Loop (min 2, max 30 rounds)
 
-#### Codex Availability Check
-
-Before starting verification rounds, check codex availability:
-
-```bash
-# Run this check once before Phase 2 begins
-which codex &>/dev/null && [ -n "$OPENAI_API_KEY" ]
-# Exit 0 → codex available: enable dual-model verification (opus + codex)
-# Exit 1 → codex unavailable: display notice and proceed with opus-only
-```
-
-If unavailable, display: `[Phase 2] Codex unavailable — opus-only verification`
-
 ```
-Team findings ──→ opus 4.6 verification ──→ codex-exec xhigh verification (if available)
-       │                                              │
+Team findings ──→ opus 4.6 verification
+       │                  │
        └── Contradiction detected? ── YES ──→ Round N+1
                                       NO  ──→ Consensus reached → Phase 3
 ```
 
 Each round:
 1. **opus 4.6**: Deep reasoning verification — checks logical consistency, identifies gaps, challenges assumptions
-2. **codex-exec xhigh** (when available): Independent code-level verification — validates technical claims, tests feasibility
-   - If unavailable: display `[Phase 2] Round {N}: Codex unavailable, proceeding with opus verification only`
-3. **Contradiction resolution**: Reconcile divergent findings between teams and verifiers
-4. **Convergence check**: All major claims verified with no outstanding contradictions → proceed
+2. **Contradiction resolution**: Reconcile divergent findings between teams and verifiers
+3. **Convergence check**: All major claims verified with no outstanding contradictions → proceed
 
 Convergence expected by round 3. Hard stop at round 30.
 
@@ -273,8 +258,7 @@ Retrieval and reasoning are distinct cognitive operations that benefit from expl
 | Phase | Model | Rationale |
 |-------|-------|-----------|
 | Phase 1 (Research teams) | sonnet | Balanced speed/quality for parallel research |
-| Phase 2 (opus verification) | opus | Deep reasoning for cross-verification |
-| Phase 2 (codex verification) | codex xhigh | Code-level validation of technical claims |
+| Phase 2 (Verification) | opus | Deep reasoning for cross-verification |
 | Phase 3 (Synthesis) | opus | Complex multi-source reasoning and taxonomy |
 
 ## Team Prompt Templates
@@ -326,17 +310,7 @@ Round N:
           - Internal consistency (breadth ↔ depth alignment)
           - Cross-domain consistency (security ↔ architecture)
           - Evidence quality (claims without backing)
-  Step 2: codex-exec validates technical claims (when available):
-          a. Invoke: /codex-exec with findings from all teams
-          b. Prompt:  "Validate technical claims: {findings}.
-                       Check code patterns, benchmark reproducibility,
-                       dependency resolution."
-          c. Effort:  --effort xhigh
-          d. Parse:   contradictions → merge with opus findings
-          e. On timeout/error: log "[Phase 2] Round {N}: codex-exec error — {reason},
-                                continuing with opus results only"
-     If unavailable: log "[Phase 2] Round {N}: Codex unavailable, proceeding with opus verification only"
-  Step 3: Compile contradiction list
+  Step 2: Compile contradiction list
           - 0 contradictions → CONVERGED
           - >0 contradictions → feedback to relevant teams → Round N+1
 ```
@@ -417,7 +391,6 @@ This advisory is informational only and does not block execution.
 
 | Scenario | Fallback |
 |----------|----------|
-| codex-exec unavailable | opus-only verification (still min 2 rounds) |
 | Agent Teams unavailable | Standard Agent tool with R009 batching |
 | Partial team failure | Synthesize from available results, note gaps in report |
 | GitHub issue creation fails | Output report to conversation only |
@@ -428,11 +401,11 @@ Before execution:
 ```
 [Research Plan] {topic}
 ├── Phase 1: 10 teams (3 batches × 4/4/2)
-├── Phase 2: Cross-verification (2-5 rounds, opus + codex)
+├── Phase 2: Cross-verification (2-5 rounds, opus)
 ├── Phase 3: Synthesis (opus)
 └── Phase 4: Report + GitHub issue
 
-Estimated: {time} | Teams: 10 | Models: sonnet → opus → codex
+Estimated: {time} | Teams: 10 | Models: sonnet → opus
 Stopping: max 30 verification rounds, convergence at 0 contradictions
 Cost: ~$8-15 (10 teams × sonnet + opus verification)
 Execute? [Y/n]

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

@@ -8,16 +8,16 @@ version: 1.0.0
 
 # roundtable-debate
 
-발산 보존(divergence preservation)을 목표로 하는 다중 에이전트 구조화 토론 스킬. 기존 `agora`(수렴 목표)와 보완 관계.
+발산 보존(divergence preservation)을 목표로 하는 다중 에이전트 구조화 토론 스킬.
 
 ## When to Use
 
-| 상황 | 권장 스킬 |
+| 상황 | 적합 여부 |
 |------|-----------|
-| 만장일치 합의 필요 (스펙 확정) | `agora` |
-| 다각적 평가 필요 (리스크 발굴) | `roundtable-debate` |
-| 단일 결정 + 단일 답변 | `agora` |
-| 의도적 발산 + 소수의견 보존 | `roundtable-debate` |
+| 다각적 평가 필요 (리스크 발굴) | 적합 |
+| 의도적 발산 + 소수의견 보존 | 적합 |
+| 단일 결정 + 단일 답변 | 부적합 (단일 에이전트 사용) |
+| 만장일치 합의 필요 (스펙 확정) | 부적합 (adversarial-review 참고) |
 
 ## Anti-Groupthink Mechanisms
 
@@ -97,7 +97,6 @@ DO NOT use Write/Edit directly on `.claude/outputs/` — CC sensitive-path guard
 
 | 스킬 | 목표 | 종료 조건 |
 |------|------|----------|
-| `agora` | 만장일치 수렴 | 모든 LLM 동의 |
 | `roundtable-debate` | 발산 보존 | 2라운드 도달 |
 | `adversarial-review` | 공격자 1인 시각 | 단일 라운드 |
 | `evaluator-optimizer` | 평가-개선 루프 | 평가 통과 |

package/templates/.claude/skills/structured-dev-cycle/SKILL.md

@@ -93,17 +93,6 @@ A PreToolUse hook in `.claude/hooks/hooks.json` checks this marker and blocks Wr
 └── Output: Implementation complete
 ```
 
-**Codex-Exec Hybrid Option**: When entering Stage 3:
-1. Check `/tmp/.claude-env-status-*` for codex CLI availability
-2. If available AND task involves new file creation → automatically delegate scaffolding to `/codex-exec`:
-   - Display: `[Codex Hybrid] Delegating scaffolding to codex-exec...`
-   - codex-exec generates initial code (strength: fast generation)
-   - Claude expert reviews and refines codex output (strength: reasoning, quality)
-3. If unavailable → display `[Codex] Unavailable — proceeding with Claude experts directly` and proceed with standard implementation via Claude experts
-
-Suitable for codex hybrid: new files, boilerplate, test stubs, scaffolding
-Not suitable: modifying existing code, architecture-dependent changes
-
 **Exit criteria**: All planned files created/modified, tests written.
 
 ### Stage 4: Verify Implementation
@@ -157,7 +146,7 @@ The stage marker file (`/tmp/.claude-dev-stage`) is read by a PreToolUse hook th
 For complex tasks, Agent Teams is **preferred** when available (R018):
 - Plan: architect agent
 - Verify: reviewer agent(s) — multi-model-verification via Agent Teams
-- Implement: domain expert agent (+ codex-exec hybrid if available)
+- Implement: domain expert agent
 - Compound: QA agent
 
 When Agent Teams is enabled AND task involves 3+ agents or review→fix cycles, using Agent Teams is MANDATORY per R018.

package/templates/CLAUDE.md

@@ -103,7 +103,7 @@ oh-my-customcode로 구동됩니다.
 | 개발 | `/dev-review`, `/dev-refactor`, `/sdd-dev`, `/structured-dev-cycle` | 코드 리뷰, 리팩토링, SDD |
 | 검증 | `/omcustom:sauron-watch`, `/deep-verify`, `/adversarial-review` | R017 검증, 릴리즈 품질, 보안 리뷰 |
 | 릴리즈 | `/pipeline auto-dev`, `/omcustom-release-notes`, `/release-plan` | 자동 개발, 릴리즈 노트 |
-| 리서치 | `/research`, `/scout`, `/deep-plan`, `/omcustom:agora` | 병렬 분석, URL 평가, 연구 계획 |
+| 리서치 | `/research`, `/scout`, `/deep-plan` | 병렬 분석, URL 평가, 연구 계획 |
 | 최적화 | `/token-efficiency-audit` | 토큰 효율 감사 (5계층 방어 스택) |
 | 시스템 | `/omcustom:lists`, `/omcustom:status`, `/omcustom:help` | 전체 목록, 상태, 도움말 |
 
@@ -116,7 +116,7 @@ project/
 +-- CLAUDE.md                    # 진입점
 +-- .claude/
 |   +-- agents/                  # 서브에이전트 정의 (49 파일)
-|   +-- skills/                  # 스킬 (118 디렉토리)
+|   +-- skills/                  # 스킬 (115 디렉토리)
 |   +-- rules/                   # 전역 규칙 (R000-R023)
 |   +-- hooks/                   # 훅 스크립트 (보안, 검증, HUD)
 |   +-- contexts/                # 컨텍스트 파일 (ecomode)
@@ -235,6 +235,7 @@ Claude Code의 Agent Teams 기능이 활성화되어 있으면 (`CLAUDE_CODE_EXP
 | cc-token-saver | ww-w-ai/cc-token-saver | 토큰/비용 최적화 (Token Guardian, /continue, /usage-view) |
 | caveman | JuliusBrussee/caveman | 영어 출력 토큰 압축 (코드 리뷰·커밋 메시지). R013 ecomode 미적용 구간 보완. 한국어 컨텍스트(R000)에는 ecomode 우선. `lite`/`full` 권장, `ultra`/`文言文` 가독성 저하 주의 |
 | cc-roundtable | gaebalai/cc-roundtable | Devil's Advocate + 소수의견 보호 + 2라운드 하드캡 다중 에이전트 토론. `roundtable-debate` 스킬에 패턴 내재화됨; 외부 플러그인 직접 사용 시 대안 경로 |
+| codex-plugin-cc | openai/codex-plugin-cc | OpenAI Codex 하이브리드 실행 — codex-exec 스킬을 대체하는 네이티브 플러그인 경로 |
 
 ### 권장 MCP 서버
 

package/templates/guides/agent-design/google-agents-cli-patterns.md

@@ -46,7 +46,7 @@ omcustom 관련성: 현재 skills는 Claude Code 전용으로 설계되어 있
 
 여러 에이전트의 출력을 단일 synthesizer가 통합하는 구조.
 
-omcustom에 이미 구현됨: `agora` skill, `result-aggregation` skill. 추가 작업 불필요.
+omcustom에 이미 구현됨: `result-aggregation` skill. 추가 작업 불필요.
 
 ## Why Not Direct Adoption
 
@@ -72,5 +72,4 @@ omcustom은 로컬 Claude Code 중심 철학 → 종속 컴포넌트 직접 채
 
 - #971 (source issue)
 - mgr-creator 에이전트 (`.claude/agents/mgr-creator.md`)
-- agora skill
 - result-aggregation skill

package/templates/guides/multi-agent-debate-patterns/README.md

@@ -25,22 +25,19 @@
 - 1명만 주장하는 의견도 별도 트랙으로 보존
 - 기각 시 명시적 정당화(3개 근거 이상) 필수
 
-## agora vs roundtable-debate 선택 매트릭스
+## 스킬 선택 매트릭스
 
 | 상황 | 권장 | 이유 |
 |------|------|------|
-| 스펙 최종 확정 (단일 답변 필요) | `agora` | 만장일치 수렴 |
 | 아키텍처 결정 (트레이드오프 평가) | `roundtable-debate` | 발산 보존 |
 | 보안 감사 (공격자 시각) | `adversarial-review` | 공격자 1인 시각 |
 | 코드 품질 개선 루프 | `evaluator-optimizer` | 평가-개선 |
 | 리스크 발굴 (블라인드 스폿) | `roundtable-debate` | 다양한 페르소나 |
-| 산출물 검증 (통과/실패) | `agora` | 단일 결정 |
 
 ## 종료 조건 비교
 
 | 스킬 | 종료 조건 | 산출물 |
 |------|----------|--------|
-| `agora` | 모든 LLM 동의 | 단일 합의안 |
 | `roundtable-debate` | 2라운드 도달 | 합의 + 소수의견 + Devil's Advocate 반대 |
 | `adversarial-review` | 단일 라운드 완료 | 취약점 목록 |
 | `evaluator-optimizer` | 평가 통과 | 개선된 산출물 |
@@ -55,11 +52,6 @@
 
 ## 구현 예시
 
-`agora`(수렴):
-```
-사용자: "이 API 스펙이 안전한가?" → agora → "안전함 (3 LLM 만장일치)"
-```
-
 `roundtable-debate`(발산):
 ```
 사용자: "OAuth vs JWT 어느 쪽이 적합?" → roundtable-debate →
@@ -75,5 +67,4 @@ cc-roundtable (https://github.com/gaebalai/cc-roundtable) 패턴에서 차용. R
 ## 관련 자료
 
 - 스킬: `.claude/skills/roundtable-debate/SKILL.md`
-- 스킬: `.claude/skills/agora/SKILL.md`
 - 룰: R009 (병렬 실행), R018 (Agent Teams)

package/templates/guides/multi-provider-exec/README.md

@@ -10,8 +10,6 @@ Inspired by OpenHarness's provider profile switching pattern, adapted for oh-my-
 
 | Provider | Skill | CLI Dependency | Model | Strengths |
 |----------|-------|---------------|-------|-----------|
-| OpenAI (Codex) | `codex-exec` | `codex` CLI | GPT-5.4 | Code generation, broad knowledge |
-| Google (Gemini) | `gemini-exec` | `gemini` CLI | Gemini 2.5 Pro | Long context, multimodal |
 | RTK (proxy) | `rtk-exec` | `rtk` CLI | Configurable | Token-optimized output, cost reduction |
 
 ## Availability Detection
@@ -20,8 +18,6 @@ The `session-env-check.sh` hook (SessionStart) auto-detects available providers:
 
 ```
 [SessionStart] Checking external CLI availability...
-  codex: ✓ available
-  gemini: ✗ not found
   rtk: ✓ available
 ```
 
@@ -32,8 +28,6 @@ Providers are opt-in — missing CLIs are silently skipped.
 ### Direct Invocation
 
 ```
-/codex-exec "Review this function for security issues"
-/gemini-exec "Analyze this architecture diagram"
 /rtk-exec "List files matching pattern X"
 ```
 
@@ -41,17 +35,12 @@ Providers are opt-in — missing CLIs are silently skipped.
 
 | Task | Recommended Provider | Rationale |
 |------|---------------------|-----------|
-| Second opinion on code review | codex-exec | Independent model reduces confirmation bias |
-| Long document analysis | gemini-exec | 1M+ context window |
 | Token-heavy batch operations | rtk-exec | Compressed output reduces context cost |
-| Security audit cross-check | codex-exec | Different training data catches different patterns |
-| Multi-model verification | All three | `/multi-model-verification` skill orchestrates this |
 
 ### Integration with Existing Skills
 
 | Skill | Uses Provider | How |
 |-------|--------------|-----|
-| `multi-model-verification` | codex-exec + gemini-exec | Parallel verification with severity classification |
 | `reasoning-sandwich` | Any exec skill | Pre/post reasoning with different models |
 | `model-escalation` | Claude models only | Internal escalation (haiku→sonnet→opus), not cross-provider |
 
@@ -60,9 +49,9 @@ Providers are opt-in — missing CLIs are silently skipped.
 | Aspect | Multi-Model Routing | Multi-Provider Exec |
 |--------|--------------------|--------------------|
 | Scope | Claude model selection | Cross-provider execution |
-| Models | haiku / sonnet / opus | GPT-5.4 / Gemini 2.5 / RTK proxy |
+| Models | haiku / sonnet / opus | RTK proxy |
 | Mechanism | `model` frontmatter field | Exec skill invocation |
-| Use case | Cost/quality optimization within Claude | Independent verification, specialized tasks |
+| Use case | Cost/quality optimization within Claude | Token-optimized output via RTK proxy |
 | Guide | `guides/multi-model-routing/` | `guides/multi-provider-exec/` |
 
 ## Configuration
@@ -71,8 +60,6 @@ No global configuration required. Each exec skill reads its own CLI configuratio
 
 | Skill | Config Source |
 |-------|-------------|
-| codex-exec | `~/.codex/config` or CODEX_API_KEY env |
-| gemini-exec | `~/.gemini/config` or GEMINI_API_KEY env |
 | rtk-exec | RTK proxy running on localhost |
 
 ## Limitations

package/templates/manifest.json

@@ -1,5 +1,5 @@
 {
-  "version": "0.158.0",
+  "version": "0.160.0",
   "lastUpdated": "2026-05-20T00:00:00.000Z",
   "omcustomMinClaudeCode": "2.1.121",
   "omcustomMinClaudeCodeReason": "Sensitive-path direct Write/Edit on .claude/** under bypassPermissions (R010 deprecation, #1101)",
@@ -20,7 +20,7 @@
       "name": "skills",
       "path": ".claude/skills",
       "description": "Reusable skill modules (includes slash commands)",
-      "files": 118
+      "files": 115
     },
     {
       "name": "guides",