ShieldClaw

import { createCore, closeCore, } from '@shieldclaw/core';
import ScanPlugin from '@shieldclaw/scan';
import GuardPlugin from '@shieldclaw/guard';
import AuditPlugin from '@shieldclaw/audit';
import VaultPlugin from '@shieldclaw/vault';

The core scanning, guard, audit, and vault behavior is delegated to external packages that are not included in the supplied artifact contents, even though those components would handle filesystem, network, logs, and secrets.

initHooks() {
        // TODO: 实现 fs/network/process 的 Hook
        this.core.logger.info('[HookFramework] Hooks initialized');
    }

The SKILL.md advertises real-time file, network, and process protection, but this visible hook implementation only logs initialization and contains a TODO for the actual hooks.

- File system monitoring
- Network request interception
- Process execution control

These are broad, high-impact controls over local and network operations. They are aligned with a security guard feature, but users should understand the authority being granted.

- "Encrypt this API Key"
- "Store this password for me"
- "Securely store this private key"
...
- System keychain for key storage

The vault feature is designed to handle credentials and private keys and to use the system keychain. This is expected for the stated purpose, but it is sensitive authority.

- **Database**: shieldclaw.db - Audit logs and configuration
- **Logs**: logs/ - Runtime logs
- **Keys**: System keychain (auto-managed)

The skill persists security logs, configuration, and vault key material locally. This is disclosed and purpose-aligned, but retained security data may be sensitive.