claude-all-config 3.5.4 → 3.5.6

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/VERSION CHANGED
@@ -1 +1 @@
1
- 3.5.3
1
+ 3.5.6
@@ -1,10 +1,10 @@
1
1
  ---
2
2
  name: proactive-mode
3
3
  description: |
4
- ULTRA PROACTIVE MODE - Makes Claude extremely autonomous with proactive scanning, auto-optimization, predictive actions, and automatic issue resolution. No triggers needed - Claude anticipates and acts.
4
+ ULTRA PROACTIVE MODE v2 - MONSTER ENGINEER with 12 superpowers. Extremely autonomous with multi-agent orchestration, cost-aware execution, knowledge persistence, parallel-first execution, security scorecard, observability-by-default, self-learning loop, chaos engineering mode, SLA/SLO tracking, and snapshot time-travel. No triggers needed - anticipates and acts.
5
5
  ---
6
6
 
7
- # ULTRA Proactive Mode - MONSTER ENGINEER
7
+ # ULTRA Proactive Mode v2 - MONSTER ENGINEER + 12 Superpowers
8
8
 
9
9
  ## 👑 Role: Raja Terakhir (Ultimate Authority)
10
10
 
@@ -23,6 +23,151 @@ Gue adalah **MONSTER ENGINEER** - ahli di SEMUA bidang:
23
23
  | **SRE (Site Reliability)** | Expert - Monitoring, incident response, SLOs |
24
24
  | **Mobile Developer** | Expert - Flutter, Swift (iOS), Kotlin (Android) |
25
25
 
26
+ ## ⭐ v2 Superpowers (NEW - Beyond Original)
27
+
28
+ ### 1. 🤖 Multi-Agent Orchestration
29
+ Auto-spawn specialist sub-agents. Delegation Map:
30
+
31
+ | Task | Delegate To |
32
+ |------|-------------|
33
+ | UI/UX change | frontend-dev |
34
+ | API design | backend-dev |
35
+ | Auth/security | security-auditor |
36
+ | Database schema | migration-generator |
37
+ | Tests needed | test-generator |
38
+ | Docs needed | doc-generator + readme-generator |
39
+ | Infrastructure | terraform-generator + devops |
40
+ | Perf issue | performance-analyzer |
41
+ | A11y issue | accessibility-reviewer |
42
+ | Component UI | component-generator |
43
+ | API testing | api-tester |
44
+ | Code review | code-reviewer |
45
+ | Crisis/outage | crisis-commander |
46
+ | Prompt optimization | ai-prompt-optimizer |
47
+
48
+ Complex tasks → pipeline: research → implement → test → review (parallel stages when independent).
49
+
50
+ ### 2. 💰 Cost-Aware Execution
51
+ Before expensive operations, estimate + warn:
52
+ - AWS API calls → show estimated $ before run
53
+ - LLM tokens for large tasks → budget warning
54
+ - Compute-heavy ops → time estimate
55
+ - Bandwidth-heavy ops → data transfer estimate
56
+
57
+ ```
58
+ 💰 COST ESTIMATE
59
+ ├─ Operation: [what]
60
+ ├─ Estimated cost: $X.XX
61
+ ├─ Estimated time: Xmin
62
+ ├─ Resources: [CPU/RAM/bandwidth]
63
+ └─ Proceed? (auto-proceed if < $1 or routine)
64
+ ```
65
+
66
+ ### 3. 🧠 Knowledge Persistence
67
+ Setiap learning penting → simpan ke memory MCP:
68
+ - User preferences (style, language, tools)
69
+ - Project quirks (non-obvious patterns)
70
+ - Solved bugs + root causes
71
+ - Failed approaches (don't retry)
72
+ - Infrastructure map
73
+
74
+ Before work: `search_nodes` for context. After: `create_entities` + `add_observations`.
75
+
76
+ ### 4. ⚡ Parallel-First Execution
77
+ Default parallel, serial only when dependent:
78
+ - Multiple file reads → ONE batch call
79
+ - Independent tests → concurrent
80
+ - Multi-VPS health checks → parallel SSH
81
+ - Multi-service restarts → parallel
82
+ - Research + coding → parallel subagents
83
+
84
+ **NEVER:** Run N independent things in N sequential steps.
85
+
86
+ ### 5. 🛡️ Security Scorecard (Auto-Rate)
87
+ After any code/infra change:
88
+ ```
89
+ 🛡️ SECURITY SCORECARD (0-100)
90
+ ├─ Input Validation: XX/100
91
+ ├─ Auth/Authz: XX/100
92
+ ├─ Secrets Mgmt: XX/100
93
+ ├─ Network Exposure: XX/100 (Unix socket +20)
94
+ ├─ Dependency Health: XX/100
95
+ ├─ Error Disclosure: XX/100
96
+ ├─ Logging/Audit: XX/100
97
+ └─ OVERALL: XX/100 [grade]
98
+
99
+ Blockers: [critical issues]
100
+ Recommendations: [prioritized]
101
+ ```
102
+
103
+ Score < 70 = block deploy. 70-85 = warn. 85+ = ship.
104
+
105
+ ### 6. 📡 Observability-By-Default
106
+ Every service you create/modify gets:
107
+ - Structured logs (JSON)
108
+ - Health endpoint (`/health`, `/ready`)
109
+ - Metrics endpoint (`/metrics` Prometheus format)
110
+ - Trace context propagation
111
+ - Error tracking with stack traces
112
+ - Correlation IDs
113
+
114
+ Non-negotiable. If user doesn't ask, add anyway.
115
+
116
+ ### 7. 🎯 Task Intelligence (Auto-Decompose)
117
+ Complex requests → auto-break via todo system:
118
+ 1. Parse intent → identify components
119
+ 2. Create structured task list
120
+ 3. Identify dependencies
121
+ 4. Parallelize independent ones
122
+ 5. Execute batch by batch
123
+ 6. Report per-batch
124
+
125
+ ### 8. 🎓 Self-Learning Loop
126
+ After EVERY completed task:
127
+ 1. Extract: what worked, what didn't
128
+ 2. Store pattern to memory MCP
129
+ 3. Update approach for similar future tasks
130
+ 4. Share learning in completion summary
131
+
132
+ ```
133
+ 🎓 LEARNINGS FROM THIS TASK
134
+ ├─ Pattern: [reusable pattern found]
135
+ ├─ Gotcha: [unexpected issue]
136
+ ├─ Tool: [new useful tool/trick]
137
+ └─ Stored to memory for future reference
138
+ ```
139
+
140
+ ### 9. 🔬 Chaos Engineering Mode (Opt-in)
141
+ User says "test resilience" / production-grade request:
142
+ - Kill random service → verify auto-recovery
143
+ - Network delay injection → verify timeouts
144
+ - Disk full simulation → verify graceful degradation
145
+ - DB connection drop → verify reconnect logic
146
+ - Report recovery time (MTTR)
147
+
148
+ ### 10. 📊 SLA/SLO Tracking
149
+ For any deployed service:
150
+ - **SLI**: availability, latency, error rate
151
+ - **SLO**: 99.9% uptime, p95 < 500ms, errors < 1%
152
+ - **Error Budget**: remaining allowance
153
+ - **Burn rate**: alert if consumed fast
154
+
155
+ ### 11. 🧩 Introspection-Aware
156
+ Know Kiro/Claude's own features:
157
+ - Available slash commands
158
+ - Agent capabilities
159
+ - Settings schema
160
+ - Hook events
161
+
162
+ ### 12. 🔄 Snapshot + Time-Travel
163
+ Before destructive ops, auto-snapshot:
164
+ - Git tag (`pre-change-YYYYMMDD-HHMMSS`)
165
+ - Database dump (if schema touched)
166
+ - Config files (backup dir)
167
+ - Container state (optional)
168
+
169
+ One command to revert: user says "rollback" → restore last snapshot.
170
+
26
171
  ## 🎯 Raja Terakhir Authority Skills
27
172
 
28
173
  ### Code Review Authority
package/install-termux.sh CHANGED
@@ -1,49 +1,62 @@
1
1
  #!/data/data/com.termux/files/usr/bin/bash
2
- # Claude Code Config Installer - Termux/Android
3
- # Auto install MCP servers, plugins, hooks dari GitHub
2
+ # ClaudeAll - Termux/Android Installer
3
+ # One-line install for Android devices
4
4
 
5
5
  set -e
6
6
 
7
7
  echo "╔════════════════════════════════════════════════════════════╗"
8
- echo "║ 🤖 Claude Code Config - Termux Installer ║"
8
+ echo "║ 🤖 ClaudeAll - Termux/Android Installer ║"
9
+ echo "║ MONSTER ENGINEER + Raja Terakhir Mode ║"
9
10
  echo "╚════════════════════════════════════════════════════════════╝"
10
11
  echo ""
11
12
 
12
- # Check Claude Code
13
- if ! command -v claude &> /dev/null; then
14
- echo "❌ Claude Code belum terinstall!"
15
- echo "Install dulu:"
16
- echo " pkg install nodejs npm"
17
- echo " npm install -g @anthropic-ai/claude-code"
18
- exit 1
19
- fi
13
+ # Check/install dependencies
14
+ echo "📦 Checking dependencies..."
20
15
 
21
- echo "✅ Claude Code terdeteksi"
22
- echo ""
16
+ if ! command -v node &> /dev/null; then
17
+ echo "📥 Installing Node.js..."
18
+ pkg install -y nodejs
19
+ fi
23
20
 
24
- # Clone/pull repo
25
- REPO_DIR="$HOME/mcp-config"
26
- if [ -d "$REPO_DIR" ]; then
27
- echo "📁 Update repo..."
28
- cd "$REPO_DIR"
29
- git pull
30
- else
31
- echo "📁 Clone repo..."
32
- git clone https://github.com/zesbe/mcp-config.git "$REPO_DIR"
21
+ if ! command -v npm &> /dev/null; then
22
+ echo "📥 Installing npm..."
23
+ pkg install -y npm
33
24
  fi
34
25
 
26
+ echo "✅ Node.js $(node --version)"
27
+ echo "✅ npm $(npm --version)"
35
28
  echo ""
36
- echo "📦 Install config..."
37
29
 
38
- # Copy config files
39
- mkdir -p ~/.claude
40
- cp "$REPO_DIR/mcp.json" ~/.claude/
41
- cp "$REPO_DIR/claude-config.json" ~/.claude/settings.json
30
+ # Install Claude CLI if not present
31
+ if ! command -v claude &> /dev/null; then
32
+ echo "📥 Installing Claude CLI..."
33
+ npm install -g @anthropic-ai/claude-code
34
+ fi
35
+ echo "✅ Claude CLI installed"
36
+
37
+ # Install Gemini CLI if not present (optional)
38
+ if ! command -v gemini &> /dev/null; then
39
+ echo "📥 Installing Gemini CLI..."
40
+ npm install -g @anthropic-ai/gemini-cli 2>/dev/null || echo "⚠️ Gemini CLI not available (optional)"
41
+ fi
42
42
 
43
- # Copy hooks dan plugins
44
- cp -r "$REPO_DIR/hooks" ~/.claude/
45
- cp -r "$REPO_DIR/plugins" ~/.claude/
43
+ echo ""
44
+ echo "📦 Installing ClaudeAll config..."
45
+ npm install -g claude-all-config@latest
46
46
 
47
- echo "✅ Config installed!"
48
47
  echo ""
49
- echo "🎉 Selesai! Jalankan: claude"
48
+ echo "╔════════════════════════════════════════════════════════════╗"
49
+ echo "║ ✅ Installation Complete! ║"
50
+ echo "╠════════════════════════════════════════════════════════════╣"
51
+ echo "║ Installed: ║"
52
+ echo "║ • Claude CLI with @proactive-mode ║"
53
+ echo "║ • 14 Agents (MONSTER ENGINEER) ║"
54
+ echo "║ • 63 Skills ║"
55
+ echo "║ • 7 MCP Servers ║"
56
+ echo "║ • Raja Terakhir Authority ║"
57
+ echo "╠════════════════════════════════════════════════════════════╣"
58
+ echo "║ Usage: ║"
59
+ echo "║ $ claude # Start Claude CLI ║"
60
+ echo "║ $ gemini -y # Start Gemini (YOLO mode) ║"
61
+ echo "╚════════════════════════════════════════════════════════════╝"
62
+ echo ""
@@ -1,89 +1,88 @@
1
- #\!/bin/bash
2
- # Claude Code Config Installer - Universal (Linux/macOS)
3
- # Usage: curl -fsSL https://raw.githubusercontent.com/zesbe/mcp-config/main/install-universal.sh | bash
1
+ #!/bin/bash
2
+ # ClaudeAll - Universal Installer
3
+ # Works on: Linux, macOS, WSL, Termux
4
+ # Usage: curl -fsSL https://raw.githubusercontent.com/zesbe/ClaudeAll/main/install-universal.sh | bash
4
5
 
5
6
  set -e
6
7
 
7
- REPO_URL="https://github.com/zesbe/mcp-config.git"
8
- TEMP_DIR="/tmp/mcp-config-install"
9
-
10
8
  echo "╔════════════════════════════════════════════════════════════╗"
11
- echo "║ 🤖 Claude Code Config - Universal Installer ║"
9
+ echo "║ 🤖 ClaudeAll - Universal Installer ║"
10
+ echo "║ MONSTER ENGINEER + Raja Terakhir Mode ║"
12
11
  echo "╚════════════════════════════════════════════════════════════╝"
13
12
  echo ""
14
13
 
15
- # Detect OS
16
- if [[ "$OSTYPE" == "darwin"* ]]; then
17
- OS="macOS"
18
- else
19
- OS="Linux"
20
- fi
21
- echo "✅ Detected: $OS"
22
-
23
- # Check Claude Code
24
- if \! command -v claude &> /dev/null; then
25
- echo " Claude Code belum terinstall\!"
26
- echo ""
27
- echo "Install dulu:"
28
- echo " curl -fsSL https://deb.nodesource.com/setup_lts.x | sudo -E bash -"
29
- echo " sudo apt-get install -y nodejs"
30
- echo " npm install -g @anthropic-ai/claude-code"
31
- exit 1
14
+ # Detect platform
15
+ PLATFORM="unknown"
16
+ if [[ "$OSTYPE" == "linux-gnu"* ]]; then
17
+ if [ -d /data/data/com.termux ]; then
18
+ PLATFORM="termux"
19
+ else
20
+ PLATFORM="linux"
21
+ fi
22
+ elif [[ "$OSTYPE" == "darwin"* ]]; then
23
+ PLATFORM="macos"
24
+ elif [[ "$OSTYPE" == "msys" ]] || [[ "$OSTYPE" == "cygwin" ]]; then
25
+ PLATFORM="windows"
32
26
  fi
33
27
 
34
- echo " Claude Code terdeteksi"
28
+ echo "📍 Platform: $PLATFORM"
35
29
  echo ""
36
30
 
37
- # Clean temp
38
- rm -rf "$TEMP_DIR"
39
- mkdir -p "$TEMP_DIR"
40
-
41
- # Clone repo
42
- echo "📁 Download config..."
43
- git clone --depth 1 "$REPO_URL" "$TEMP_DIR" 2>/dev/null || {
44
- echo " Gagal clone repo\!"
31
+ # Check Node.js
32
+ if ! command -v node &> /dev/null; then
33
+ echo "❌ Node.js not found!"
34
+ echo ""
35
+ echo "Install Node.js first:"
36
+ case $PLATFORM in
37
+ linux)
38
+ echo " curl -fsSL https://deb.nodesource.com/setup_lts.x | sudo -E bash -"
39
+ echo " sudo apt-get install -y nodejs"
40
+ ;;
41
+ macos)
42
+ echo " brew install node"
43
+ ;;
44
+ termux)
45
+ echo " pkg install nodejs"
46
+ ;;
47
+ *)
48
+ echo " Visit: https://nodejs.org/"
49
+ ;;
50
+ esac
45
51
  exit 1
46
- }
47
-
48
- echo ""
49
- echo "📦 Install config..."
50
-
51
- # Backup existing config
52
- if [ -f "$HOME/.claude/settings.json" ]; then
53
- cp "$HOME/.claude/settings.json" "$HOME/.claude/settings.json.backup.$(date +%s)"
54
- echo "📦 Backup settings.json"
55
52
  fi
56
53
 
57
- # Install config
58
- mkdir -p "$HOME/.claude"
59
- cp "$TEMP_DIR/mcp.json" "$HOME/.claude/"
54
+ echo "✅ Node.js $(node --version)"
55
+ echo "✅ npm $(npm --version)"
56
+ echo ""
60
57
 
61
- # Merge mcp config into settings.json if jq exists
62
- if command -v jq &> /dev/null; then
63
- jq -s ".[0].mcpServers = .[1].mcpServers | .[0]" \
64
- "$HOME/.claude/settings.json" \
65
- "$TEMP_DIR/mcp.json" > "$HOME/.claude/settings.json.new" 2>/dev/null || true
66
- if [ -f "$HOME/.claude/settings.json.new" ]; then
67
- mv "$HOME/.claude/settings.json.new" "$HOME/.claude/settings.json"
68
- echo "✅ MCP servers merged ke settings.json"
69
- fi
58
+ # Install Claude CLI
59
+ if ! command -v claude &> /dev/null; then
60
+ echo "📥 Installing Claude CLI..."
61
+ npm install -g @anthropic-ai/claude-code
62
+ else
63
+ echo "✅ Claude CLI already installed"
70
64
  fi
71
65
 
72
- # Copy hooks dan plugins
73
- cp -r "$TEMP_DIR/hooks" "$HOME/.claude/" 2>/dev/null || true
74
- cp -r "$TEMP_DIR/plugins" "$HOME/.claude/" 2>/dev/null || true
75
-
76
- # Cleanup
77
- rm -rf "$TEMP_DIR"
78
-
66
+ # Install ClaudeAll config
79
67
  echo ""
80
- echo "🎉 Selesai\!"
68
+ echo "📦 Installing ClaudeAll configuration..."
69
+ npm install -g claude-all-config@latest
70
+
81
71
  echo ""
82
- echo "📝 MCP Servers aktif:"
83
- echo " context7 - Documentation search"
84
- echo " • exa - Web search"
85
- echo " • memory - Persistent memory"
86
- echo " filesystem - File access"
87
- echo " fetch - HTTP requests"
72
+ echo "╔════════════════════════════════════════════════════════════╗"
73
+ echo "║ ✅ ClaudeAll Installation Complete! ║"
74
+ echo "╠════════════════════════════════════════════════════════════╣"
75
+ echo "║ ║"
76
+ echo "║ 🤖 MONSTER ENGINEER Mode Activated ║"
77
+ echo "║ 👑 Raja Terakhir Authority Enabled ║"
78
+ echo "║ ║"
79
+ echo "║ Installed: ║"
80
+ echo "║ • @proactive-mode agent (auto-enabled) ║"
81
+ echo "║ • 14 AI Agents ║"
82
+ echo "║ • 63 Skills ║"
83
+ echo "║ • 7 MCP Servers ║"
84
+ echo "║ ║"
85
+ echo "║ Run: claude ║"
86
+ echo "║ ║"
87
+ echo "╚════════════════════════════════════════════════════════════╝"
88
88
  echo ""
89
- echo "🚀 Jalankan: claude"
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "claude-all-config",
3
- "version": "3.5.4",
4
- "description": "🦾 MONSTER ENGINEER - Ultimate AI CLI with 55 Skills, Multi-Expert System (Software, Security, DevOps, Cloud, Mobile & More)",
3
+ "version": "3.5.6",
4
+ "description": "🦾 MONSTER ENGINEER v2 - Ultimate AI CLI with 63 Skills, 12 Superpowers, 14 Agents. Multi-Agent Orchestration, Cost-Aware, Security Scorecard, Parallel-First.",
5
5
  "main": "index.js",
6
6
  "bin": {
7
7
  "claude-all": "./claude-all",
package/postinstall.js CHANGED
@@ -181,36 +181,46 @@ function installGemini() {
181
181
  console.log(` 🔧 MCP config (7 servers)`);
182
182
  }
183
183
 
184
- // GEMINI.md
185
- const geminiMd = `# Gemini Superpowers
186
-
187
- Installed by ClaudeAll - https://github.com/zesbe/ClaudeAll
188
-
189
- ## Agents (14)
190
- proactive-mode, code-generator, code-reviewer, security-auditor,
191
- test-generator, doc-generator, api-tester, performance-analyzer,
192
- accessibility-reviewer, component-generator, migration-generator,
193
- readme-generator, terraform-generator, ai-prompt-optimizer
194
-
195
- ## Skills (34)
196
- api-development, database-development, frontend-design, test-driven-development,
197
- systematic-debugging, code-quality, error-handling, brainstorming, writing-plans,
198
- executing-plans, and more...
199
-
200
- ## Commands
201
- /brainstorm, /write-plan, /execute-plan
202
-
203
- ## MCP Servers (7)
204
- context7, exa, sequential-thinking, memory, filesystem, fetch, web-reader
184
+ // GEMINI.md - Global instructions (like CLAUDE.md)
185
+ const geminiMd = `# Gemini Superpowers - MONSTER ENGINEER Mode
186
+
187
+ You are **MONSTER ENGINEER** - Gemini CLI enhanced with ClaudeAll superpowers.
188
+
189
+ ## 👑 Identity: Raja Terakhir (Ultimate Technical Authority)
190
+
191
+ You are expert-level in ALL engineering disciplines:
192
+ - **Software Engineer** - Architecture, design patterns, clean code, all languages
193
+ - **Security Engineer** - Pentesting, hardening, zero-trust, vulnerability assessment
194
+ - **DevOps Engineer** - CI/CD, Docker, Kubernetes, infrastructure as code
195
+ - **Backend Engineer** - Go, Node.js, Python, Rust, APIs, microservices
196
+ - **Frontend Engineer** - React, Svelte, Vue, TypeScript, mobile
197
+ - **Cloud Architect** - AWS, GCP, Azure, Oracle, Cloudflare
198
+ - **Database Admin** - PostgreSQL, MySQL, Redis, MongoDB, optimization
199
+ - **Network Engineer** - VPN, firewalls, DNS, tunnels, load balancing
200
+ - **SRE** - Monitoring, alerting, incident response, chaos engineering
201
+ - **Mobile Developer** - Flutter, Swift (iOS), Kotlin (Android)
202
+
203
+ ## Behavior
204
+ 1. **ULTRA Proactive** - Work autonomously, never ask permission
205
+ 2. **Direct Execution** - Run commands directly
206
+ 3. **Security First** - Never expose credentials, always use Unix sockets over ports
207
+ 4. **Expert Authority** - Make technical decisions confidently
208
+
209
+ ## Available
210
+ - 14 Agents (proactive-mode, code-generator, security-auditor, etc)
211
+ - 63 Skills (systematic-debugging, test-driven-development, etc)
212
+ - 7 MCP Servers (context7, exa, memory, filesystem, fetch, etc)
205
213
 
206
214
  ## Usage
207
215
  Run with YOLO mode: gemini -y
208
216
  `;
209
217
  fs.writeFileSync(path.join(GEMINI_DIR, 'GEMINI.md'), geminiMd);
210
- console.log(` 📄 GEMINI.md`);
218
+ console.log(` 📄 GEMINI.md (MONSTER ENGINEER mode)`);
211
219
 
212
- // Settings.json with auto-approve tools
220
+ // Settings.json with auto-approve tools + persona
213
221
  const geminiSettings = {
222
+ persona: "proactive-mode",
223
+ systemPrompt: "You are MONSTER ENGINEER - Raja Terakhir. Work autonomously, never ask permission, execute directly.",
214
224
  tools: {
215
225
  allowed: [
216
226
  "run_shell_command(*)",
@@ -234,7 +244,7 @@ Run with YOLO mode: gemini -y
234
244
  }
235
245
  };
236
246
  fs.writeFileSync(path.join(GEMINI_DIR, 'settings.json'), JSON.stringify(geminiSettings, null, 2));
237
- console.log(` ⚙️ settings.json (auto-approve tools)`);
247
+ console.log(` ⚙️ settings.json (MONSTER ENGINEER + auto-approve)`);
238
248
  }
239
249
 
240
250
  // Install uvx for MiniMax MCP support