oh-my-customcode 0.19.0 → 0.19.4

package/README.md

@@ -124,7 +124,7 @@ Claude Code selects the appropriate model and parallelizes independent tasks (up
 | **QA** | 3 | qa-planner, qa-writer, qa-engineer |
 | **Total** | **41** | |
 
-### Skills (55)
+### Skills (56)
 
 | Category | Count | Skills |
 |----------|-------|--------|

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "oh-my-customcode",
-  "version": "0.19.0",
+  "version": "0.19.4",
   "description": "Batteries-included agent harness for Claude Code",
   "type": "module",
   "bin": {

package/templates/.claude/hooks/scripts/session-env-check.sh

@@ -27,16 +27,34 @@ if [ "${CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS:-0}" = "1" ]; then
   AGENT_TEAMS_STATUS="enabled"
 fi
 
+# Git workflow reminder
+CURRENT_BRANCH="unknown"
+if command -v git >/dev/null 2>&1 && git rev-parse --is-inside-work-tree >/dev/null 2>&1; then
+  CURRENT_BRANCH=$(git branch --show-current 2>/dev/null || echo "unknown")
+fi
+
 # Write status to file for other hooks to reference
 STATUS_FILE="/tmp/.claude-env-status-${PPID}"
 cat > "$STATUS_FILE" << ENVEOF
 codex=${CODEX_STATUS}
 agent_teams=${AGENT_TEAMS_STATUS}
+git_branch=${CURRENT_BRANCH}
 ENVEOF
 
 # Report to stderr (visible in conversation)
 echo "  codex CLI: ${CODEX_STATUS}" >&2
 echo "  Agent Teams: ${AGENT_TEAMS_STATUS}" >&2
+echo "" >&2
+echo "  [Git Workflow Reminder]" >&2
+echo "  Current branch: ${CURRENT_BRANCH}" >&2
+if [ "$CURRENT_BRANCH" = "develop" ] || [ "$CURRENT_BRANCH" = "main" ] || [ "$CURRENT_BRANCH" = "master" ]; then
+  echo "  ⚠ You are on a protected branch!" >&2
+  echo "  ⚠ Create a feature branch before making changes:" >&2
+  echo "    git checkout -b feat/your-feature develop" >&2
+else
+  echo "  ✓ Feature branch detected" >&2
+fi
+echo "  Rules: feature branch → commit → push → PR → merge" >&2
 echo "------------------------------------" >&2
 
 # Pass through

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

@@ -48,6 +48,28 @@ BEFORE using Task tool for 2+ agent tasks, this check is **ENFORCED**:
 ╚══════════════════════════════════════════════════════════════════╝
 ```
 
+### Spawn Completeness Check (MANDATORY)
+
+When spawning Agent Teams members:
+
+**ALL members MUST be spawned in a SINGLE message.** Partial spawning is a VIOLATION of both R018 and R009.
+
+```
+╔══════════════════════════════════════════════════════════════════╗
+║  BEFORE SPAWNING TEAM MEMBERS:                                   ║
+║                                                                   ║
+║  1. How many members does this team need?  N = ___               ║
+║  2. Am I spawning ALL N members in THIS message?                 ║
+║     YES → Good. Continue.                                        ║
+║     NO  → STOP. This is a VIOLATION.                             ║
+║           All N members MUST be in the same message.             ║
+║                                                                   ║
+║  Partial spawn (e.g., 1/3) = VIOLATION                           ║
+║  Sequential spawn (one per message) = VIOLATION                  ║
+║  All at once in single message = CORRECT                         ║
+╚══════════════════════════════════════════════════════════════════╝
+```
+
 ## Common Violations
 
 ```
@@ -87,6 +109,19 @@ BEFORE using Task tool for 2+ agent tasks, this check is **ENFORCED**:
    Task(frontend-dev) + Task(backend-dev) → team members
    Shared TaskList for interface contracts
    SendMessage for API schema coordination
+
+❌ WRONG: Spawning team members one at a time
+   TeamCreate("research-team")
+   Message 1: Task(researcher-1) → Analysis 1   (only 1/3 spawned)
+   Message 2: Task(researcher-2) → Analysis 2   (late spawn)
+   Message 3: Task(researcher-3) → Analysis 3   (late spawn)
+
+✓ CORRECT: All members in a single message
+   TeamCreate("research-team")
+   Single message:
+     Task(researcher-1) → Analysis 1  ┐
+     Task(researcher-2) → Analysis 2  ├─ ALL spawned together
+     Task(researcher-3) → Analysis 3  ┘
 ```
 
 ## Cost Guidelines

package/templates/.claude/rules/MUST-parallel-execution.md

@@ -39,6 +39,7 @@ Before writing/editing multiple files:
 2. Using Write/Edit sequentially for 2+ files? → STOP, parallelize
 3. Specialized agent available? → Use it (not general-purpose)
 4. Agent Teams available + 3+ agents or review cycle? → YES: use Agent Teams instead of Task
+5. Agent Teams members? → ALL members MUST spawn in a single message (no partial spawning)
 
 ### Common Violations to Avoid
 
@@ -79,6 +80,18 @@ Before writing/editing multiple files:
    Task(lang-kotlin-expert → usecase queries)     ├─ All spawned together
    Task(be-springboot-expert → persistence)       │
    Task(be-springboot-expert → security)          ┘
+
+❌ WRONG: Agent Teams partial spawn (1 of N members)
+   TeamCreate("feature-team")
+   Message 1: Task(member-1)  ← only 1/3 spawned, VIOLATION
+   Message 2: Task(member-2)  ← sequential, VIOLATION
+   Message 3: Task(member-3)  ← sequential, VIOLATION
+
+✓ CORRECT: All Agent Teams members in single message
+   TeamCreate("feature-team")
+   Task(member-1)  ┐
+   Task(member-2)  ├─ Single message, all at once
+   Task(member-3)  ┘
 ```
 
 ## Execution Rules

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

@@ -0,0 +1,161 @@
+---
+name: analysis
+description: Analyze project and auto-configure agents, skills, rules, and guides
+argument-hint: "[--dry-run] [--verbose]"
+---
+
+# Project Analysis Skill
+
+Scan a project's tech stack, compare against installed agents/skills, and auto-configure missing items.
+
+## Options
+
+```
+--dry-run    Show what would be added without making changes
+--verbose    Show detailed detection reasoning
+```
+
+## Workflow
+
+### Step 1: Project Scan
+
+Detect tech stack by checking indicator files and dependency manifests.
+
+| Indicator | Files to Check | Agent | Skill |
+|-----------|---------------|-------|-------|
+| TypeScript | tsconfig.json, *.ts, *.tsx | lang-typescript-expert | typescript-best-practices |
+| React/Next.js | next.config.*, package.json (next dep) | fe-vercel-agent | react-best-practices |
+| Vue.js | vue.config.*, *.vue | fe-vuejs-agent | - |
+| Svelte | svelte.config.*, *.svelte | fe-svelte-agent | - |
+| Python | pyproject.toml, requirements.txt, *.py | lang-python-expert | python-best-practices |
+| FastAPI | "fastapi" in imports/deps | be-fastapi-expert | fastapi-best-practices |
+| Go | go.mod, *.go | lang-golang-expert | go-best-practices |
+| Go Backend | go.mod + cmd/ or internal/ dirs | be-go-backend-expert | go-backend-best-practices |
+| Rust | Cargo.toml, *.rs | lang-rust-expert | rust-best-practices |
+| Kotlin | *.kt, build.gradle.kts | lang-kotlin-expert | kotlin-best-practices |
+| Java | *.java, pom.xml | lang-java21-expert | - |
+| Spring Boot | spring-boot in deps | be-springboot-expert | springboot-best-practices |
+| Express.js | "express" in deps | be-express-expert | - |
+| NestJS | "@nestjs" in deps | be-nestjs-expert | - |
+| Docker | Dockerfile, compose.yml | infra-docker-expert | docker-best-practices |
+| AWS | CDK/SAM/CloudFormation files | infra-aws-expert | aws-best-practices |
+| PostgreSQL | *.sql, pg in deps | db-postgres-expert | postgres-best-practices |
+| Redis | redis in deps | db-redis-expert | redis-best-practices |
+| Supabase | supabase in deps/config | db-supabase-expert | supabase-postgres-best-practices |
+| Airflow | dags/*.py, airflow in deps | de-airflow-expert | airflow-best-practices |
+| dbt | dbt_project.yml | de-dbt-expert | dbt-best-practices |
+| Kafka | kafka in deps/config | de-kafka-expert | kafka-best-practices |
+| Spark | spark in deps/config | de-spark-expert | spark-best-practices |
+| Snowflake | snowflake in deps/config | de-snowflake-expert | snowflake-best-practices |
+
+**Detection logic:**
+
+```
+1. Read package.json / go.mod / Cargo.toml / pyproject.toml / pom.xml
+2. Glob for indicator files (tsconfig.json, *.vue, Dockerfile, etc.)
+3. Grep dependencies for framework/library names
+4. For verbose mode: log each indicator found and confidence level
+```
+
+### Step 2: Gap Analysis
+
+Compare detected stack against what is already installed.
+
+```
+1. List existing agents:  ls .claude/agents/*.md
+2. List existing skills:  find .claude/skills -name "SKILL.md"
+3. For each detected indicator:
+   a. Check if required agent file exists → mark MISSING or PRESENT
+   b. Check if required skill directory exists → mark MISSING or PRESENT
+4. Build two lists:
+   - missing_agents[]   — agents needed but not present
+   - missing_skills[]   — skills needed but not present
+5. (Optional) Build unused list for suggestions:
+   - Agents present but no indicator matched → flag for review
+```
+
+### Step 3: Auto-Configure
+
+Apply changes for all missing items (skip in --dry-run mode).
+
+```
+For each missing agent:
+  - If agent exists in templates/.claude/agents/ → copy to .claude/agents/
+  - Else → delegate to mgr-creator with detected domain context
+
+For each missing skill:
+  - If skill exists in templates/.claude/skills/ → copy to .claude/skills/
+  - Else → log as "skill not available in templates, manual setup needed"
+
+Rules:
+  - Keep all existing rules (they are universal, never remove)
+
+Guides:
+  - Verify guides/ directory has relevant reference docs
+  - Log missing guide topics as suggestions only (no auto-copy)
+```
+
+### Step 4: Report
+
+Output a structured summary after the run.
+
+```
+[analysis] Project: <detected project name or path>
+
+Tech Stack Detected:
+  - TypeScript (tsconfig.json found)
+  - React/Next.js (next in package.json deps)
+  - Docker (Dockerfile found)
+
+Agents:
+  + lang-typescript-expert  [added]
+  + fe-vercel-agent          [added]
+  ~ infra-docker-expert      [already present, skipped]
+
+Skills:
+  + typescript-best-practices  [added]
+  + react-best-practices        [added]
+  ~ docker-best-practices       [already present, skipped]
+
+Rules:   no changes (universal rules kept as-is)
+
+Guides:  react/ — present
+         docker/ — present
+         typescript/ — present
+
+Suggestions:
+  - infra-aws-expert not detected (no CDK/SAM files found)
+  - de-* agents not detected (no pipeline indicators found)
+
+Summary: 2 agents added, 2 skills added, 0 removed
+```
+
+**--dry-run output** prefixes all additions with `[would add]` instead of `[added]` and makes no file changes.
+
+**--verbose output** adds a Detection section before the report:
+
+```
+Detection Details:
+  tsconfig.json            → TypeScript confirmed
+  package.json[next]       → Next.js confirmed (confidence: high)
+  package.json[react]      → React confirmed (confidence: high)
+  Dockerfile               → Docker confirmed
+  no go.mod found          → Go skipped
+  no Cargo.toml found      → Rust skipped
+```
+
+## Example Invocation
+
+```
+/analysis
+/analysis --dry-run
+/analysis --verbose
+/analysis --dry-run --verbose
+```
+
+## Notes
+
+- Always run `--dry-run` first on unfamiliar projects to preview changes
+- Agents and skills are additive only — existing items are never removed automatically
+- For stacks not in the detection table, delegate to `mgr-creator` for dynamic agent creation
+- Rules are never auto-removed; they are project-universal

package/templates/CLAUDE.md.en

@@ -145,6 +145,7 @@ Violation = immediate correction. No exception for "small changes".
 
 | Command | Description |
 |---------|-------------|
+| `/analysis` | Analyze project and auto-configure customizations |
 | `/create-agent` | Create a new agent |
 | `/update-docs` | Sync documentation with project structure |
 | `/update-external` | Update agents from external sources |
@@ -174,7 +175,7 @@ project/
 +-- CLAUDE.md                    # Entry point
 +-- .claude/
 |   +-- agents/                  # Subagent definitions (41 files)
-|   +-- skills/                  # Skills (55 directories)
+|   +-- skills/                  # Skills (56 directories)
 |   +-- rules/                   # Global rules (R000-R018)
 |   +-- hooks/                   # Hook scripts (memory, HUD)
 |   +-- contexts/                # Context files (ecomode)
@@ -238,6 +239,9 @@ Task tool + routing skills remain the fallback for simple/cost-sensitive tasks.
 ## Quick Reference
 
 ```bash
+# Project analysis
+/analysis
+
 # Show all commands
 /lists
 

package/templates/CLAUDE.md.ko

@@ -145,6 +145,7 @@ oh-my-customcode로 구동됩니다.
 
 | 커맨드 | 설명 |
 |--------|------|
+| `/analysis` | 프로젝트 분석 및 자동 커스터마이징 |
 | `/create-agent` | 새 에이전트 생성 |
 | `/update-docs` | 프로젝트 구조와 문서 동기화 |
 | `/update-external` | 외부 소스에서 에이전트 업데이트 |
@@ -174,7 +175,7 @@ project/
 +-- CLAUDE.md                    # 진입점
 +-- .claude/
 |   +-- agents/                  # 서브에이전트 정의 (41 파일)
-|   +-- skills/                  # 스킬 (55 디렉토리)
+|   +-- skills/                  # 스킬 (56 디렉토리)
 |   +-- rules/                   # 전역 규칙 (R000-R018)
 |   +-- hooks/                   # 훅 스크립트 (메모리, HUD)
 |   +-- contexts/                # 컨텍스트 파일 (ecomode)
@@ -238,6 +239,9 @@ Claude Code의 Agent Teams 기능이 활성화되어 있으면 (`CLAUDE_CODE_EXP
 ## 빠른 참조
 
 ```bash
+# 프로젝트 분석
+/analysis
+
 # 모든 커맨드 표시
 /lists
 

package/templates/manifest.json

@@ -18,7 +18,7 @@
       "name": "skills",
       "path": ".claude/skills",
       "description": "Reusable skill modules (includes slash commands)",
-      "files": 55
+      "files": 56
     },
     {
       "name": "guides",