ClawHub
Back to skill

Security audit

CCPA Audit - CCPA合规审计

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a disclosed CCPA audit tool, but its advertised offline preview still contacts the cloud service, and it handles account credentials and stored API keys.

Review before installing if you need a truly local tool: the preview mode is documented as offline but the code may still contact compliancehub.cn to fetch rules. Use a separate password for the optional account flow, understand that scored audits send your answers and API key to the service, and delete ~/.config/compliancehub/ccpa-audit.key if you no longer want the key stored locally.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Low
Confidence
84% confidence
Finding
The trigger list includes broad privacy-related phrases such as '隐私合规报告', which could cause the skill to activate for generic privacy/compliance requests outside a narrow CCPA scope. This can route users into a workflow that encourages cloud submission of compliance answers and optional credential entry when they may have intended a different tool or a local-only response.

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
The package metadata presents key behavior and risk disclosures primarily in Chinese, including notices about cloud transmission of user answers and optional credential collection. Users who do not read Chinese may trigger networked processing or credential-handling flows without meaningful informed consent, which undermines transparency for a privacy/compliance-focused skill.

Session Persistence

Medium
Category
Rogue Agent
Content
2. If the user wants a scored report, prompt them to get a free Key (`--login` or the account page), then run the full 审计.

## Security & data handling
- **Account credentials (only on explicit `--login`):** When you run `--login`/`--auth`, the skill prompts for your email + password and POSTs them ONLY to compliancehub.cn's official auth endpoints (/api/v1/auth/register, /api/v1/auth/login) to create your account and issue the free API Key.
- **Where data goes:** Check items are fetched from, and your answers are scored by, the CQDev cloud at
  `https://compliancehub.cn` (the operator's official endpoint). Scoring transmits only your item answers and the free API Key (as a Bearer token).
- **API Key storage (user-initiated only):** written **only when you run `--login`** to
Confidence
93% confidence
Finding
create your account and issue the free API Key. - **Where data goes:** Check items are fetched from, and your answers are scored by, the CQDev cloud at `https://compliancehub.cn` (the operator's off

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.