ClawHub
Back to skill

Security audit

cmclaw

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed YouCloud Creative Manager integration that sends user prompts to YouCloud for ad strategy analysis, with no evidence of hidden or destructive behavior.

Install this only if you intend to use YouCloud Creative Manager and are comfortable sending ad strategy prompts, material-scope references, and account-linked user/team identifiers to YouCloud, where conversations may appear in the web app's AI strategy conversation list. Keep YOUCLOUD_API_KEY scoped to the intended account and avoid setting DAM_API_BASE to an untrusted host.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (8)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill declares access to environment-backed credentials and performs networked API calls, but it does not declare an explicit tool/permission scope that constrains those capabilities. In practice, this increases the chance of overbroad execution or unauthorized external calls because the runtime and reviewers cannot clearly enforce least-privilege boundaries from the manifest alone.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger phrases are broad enough to match common requests such as brainstorming, inspiration, or strategy analysis, which can cause the skill to activate when the user did not intend to call it. Because the skill then performs authenticated remote API interactions and may persist session context, accidental invocation can expose user prompts and metadata to an external service unexpectedly.

Description-Behavior Mismatch

Medium
Confidence
89% confidence
Finding
The API documentation explicitly states that skill-initiated conversations are synchronized to an external web application's conversation list. That creates an external data-sharing side effect beyond simple local analysis/report generation and may expose user prompts, uploaded material descriptions, or business strategy content to another product surface without clear user consent. In a marketing-strategy skill, those prompts can contain sensitive campaign plans or proprietary creative material, making the side effect materially relevant.

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
The file is written as a prescriptive skill API reference entirely in Chinese and does not indicate that users or operators may choose another language or locale. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation.

External Transmission

Medium
Category
Data Exfiltration
Content
```bash
# 环境变量：YOUCLOUD_API_KEY / DAM_USER_ID / DAM_TEAM_ID（后两者由 userInfo:get 返回）
curl -N -X POST https://console.dam.youcloud.com/api/rpc/ai/claw/v1/chat \
  -H "Authorization: YC_API_KEY $YOUCLOUD_API_KEY" \
  -H "Content-Type: application/json; charset=utf-8" \
  --max-time 650 \
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
```python
import json, requests

resp = requests.post(
    "https://console.dam.youcloud.com/api/rpc/ai/claw/v1/chat",
    headers={"Authorization": f"YC_API_KEY {api_key}"},
    json={"input": question, "user_id": int(uid), "team_id": tid, "chat_mode": 2},  # uid 转 int
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
```python
import json, requests

resp = requests.post(
    "https://console.dam.youcloud.com/api/rpc/ai/claw/v1/chat",
    headers={"Authorization": f"YC_API_KEY {api_key}"},
    json={"input": question, "user_id": int(uid), "team_id": tid, "chat_mode": 2},  # uid 转 int
Confidence
70% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Natural-Language Policy Violations

Low
Confidence
83% confidence
Finding
The user-facing operational guidance and required prompts are specified only in Chinese, including the mandated range-selection text and workflow instructions. The file does not state that the skill is limited to Chinese-speaking users or offer any user opt-in or language-selection behavior, which can violate language/locale policy expectations.

Static analysis

No suspicious patterns detected.