aiblueprint-cli 1.1.8 → 1.2.1

package/README.md

@@ -517,6 +517,44 @@ bun run dev claude-code setup --folder ./test-config
 3. Test installation: `bun run dev claude-code add commands <name>`
 4. Document in README
 
+## 💎 AIBlueprint CLI Premium
+
+Unlock advanced features with **AIBlueprint CLI Premium**:
+
+### Premium Features
+
+- ✨ **Advanced Statusline** - Enhanced git info, real-time costs, and token usage tracking
+- 🚀 **Premium Commands** - Exclusive workflow automation templates
+- 🤖 **Premium Agents** - Specialized AI agents for complex tasks
+- 📦 **Priority Updates** - Get new features first
+- 💬 **Priority Support** - Direct help when you need it
+
+### How to Upgrade
+
+```bash
+# 1. Get your premium token at https://mlv.sh/claude-cli
+
+# 2. Activate premium
+aiblueprint claude-code pro activate YOUR_TOKEN
+
+# 3. Setup premium configs
+aiblueprint claude-code pro setup
+```
+
+### Premium Commands
+
+```bash
+# Check premium status
+aiblueprint claude-code pro status
+
+# Update premium configs
+aiblueprint claude-code pro update
+```
+
+**Learn more:** https://mlv.sh/claude-cli
+
+---
+
 ## 📄 License
 
 MIT License - see [LICENSE](LICENSE) file for details.

package/claude-code-config/scripts/command-validator/README.md

@@ -0,0 +1,147 @@
+# Command Validator
+
+A secure command validation package for Claude Code's PreToolUse hook. This package validates bash commands before execution to prevent dangerous operations.
+
+## Features
+
+- **Comprehensive Security Rules**: Blocks dangerous commands (rm -rf /, dd, mkfs, etc.)
+- **Pattern Matching**: Detects malicious patterns like fork bombs, backdoors, and data exfiltration
+- **Path Protection**: Prevents writes to system directories (/etc, /usr, /bin, etc.)
+- **Command Chaining**: Validates chained commands (&&, ||, ;)
+- **Fully Tested**: 82+ tests with Vitest ensuring reliable validation
+
+## Installation
+
+```bash
+bun install
+```
+
+## Usage
+
+### As a Claude Code Hook
+
+The validator is configured as a PreToolUse hook in Claude Code settings:
+
+```json
+{
+  "hooks": {
+    "PreToolUse": [
+      {
+        "matcher": "Bash",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "bun /Users/melvynx/.claude/scripts/command-validator/src/cli.ts"
+          }
+        ]
+      }
+    ]
+  }
+}
+```
+
+### Programmatic Usage
+
+```typescript
+import { CommandValidator } from "./src/lib/validator";
+
+const validator = new CommandValidator();
+const result = validator.validate("rm -rf /");
+
+if (!result.isValid) {
+  console.log(`Blocked: ${result.violations.join(", ")}`);
+  console.log(`Severity: ${result.severity}`);
+}
+```
+
+## Testing
+
+```bash
+# Run all tests
+bun test
+
+# Run tests with UI
+bun test:ui
+```
+
+## Test Coverage
+
+The test suite includes:
+
+### Safe Commands (Must Allow)
+- Standard utilities: ls, git, npm, pnpm, node, python
+- File operations: cat, cp, mv, mkdir, touch
+- Safe command chains with &&
+
+### Dangerous Commands (Must Block)
+- System destruction: rm -rf /, dd, mkfs, fdisk
+- Privilege escalation: sudo, chmod, chown, passwd
+- Network attacks: nc, nmap, telnet
+- Malicious patterns: fork bombs, backdoors, log manipulation
+- Sensitive file access: /etc/passwd, /etc/shadow, /etc/sudoers
+
+### Special Cases
+- rm -rf safety: Allows deletions in safe paths (/Users/melvynx/Developer/, /tmp/)
+- Protected paths: Blocks dangerous operations on /etc, /usr, /bin, etc.
+- Binary content detection
+- Command length limits
+
+## Architecture
+
+```
+src/
+├── cli.ts                 # CLI entry point (used by Claude Code hook)
+├── lib/
+│   ├── types.ts           # TypeScript interfaces
+│   ├── security-rules.ts  # Security rules database
+│   └── validator.ts       # Core validation logic
+└── __tests__/
+    └── validator.test.ts  # Comprehensive test suite
+```
+
+## Security Rules
+
+### Critical Commands
+- `del`, `format`, `mkfs`, `shred`, `dd`, `fdisk`, `parted`
+
+### Privilege Escalation
+- `sudo`, `su`, `passwd`, `chpasswd`, `usermod`, `chmod`, `chown`
+
+### Network Commands
+- `nc`, `netcat`, `nmap`, `telnet`, `ssh-keygen`, `iptables`
+
+### System Manipulation
+- `systemctl`, `service`, `kill`, `killall`, `mount`, `umount`
+
+### Protected Paths
+- `/etc/`, `/usr/`, `/sbin/`, `/boot/`, `/sys/`, `/proc/`, `/dev/`, `/root/`
+
+## Security Logs
+
+Security events are logged to `data/security.log` inside the package directory. The log file contains:
+- Timestamp
+- Session ID
+- Tool name
+- Command (truncated to 500 chars)
+- Blocked/allowed status
+- Severity level
+- Violations detected
+
+The `data/` folder is gitignored to prevent committing sensitive log data.
+
+## Development
+
+```bash
+# Run linter
+bun run lint
+
+# Format code
+bun run format
+
+# Type check
+bunx tsc --noEmit
+```
+
+## License
+
+MIT

package/claude-code-config/scripts/command-validator/biome.json

@@ -0,0 +1,29 @@
+{
+	"$schema": "https://biomejs.dev/schemas/2.3.4/schema.json",
+	"vcs": {
+		"enabled": false,
+		"clientKind": "git",
+		"useIgnoreFile": false
+	},
+	"files": {
+		"ignoreUnknown": false
+	},
+	"formatter": {
+		"enabled": true,
+		"indentStyle": "tab"
+	},
+	"linter": {
+		"enabled": true,
+		"rules": {
+			"recommended": true,
+			"suspicious": {
+				"noControlCharactersInRegex": "off"
+			}
+		}
+	},
+	"javascript": {
+		"formatter": {
+			"quoteStyle": "double"
+		}
+	}
+}