ClawHub
Back to skill

Security audit

Aloudata CAN SKILLS - metric-attribution

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly performs metric attribution as advertised, but it can send business metric context through networked services under broad triggers, including an undeclared web search path.

Review before installing if your metric names, dimensions, regions, platforms, or time windows are confidential. The Gateway API access is expected, but the skill should ideally disclose and scope the WebSearch path or require confirmation before sending business context to search providers.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (7)

Vague Triggers

High
Confidence
97% confidence
Finding
The trigger conditions are extremely broad and include many everyday phrases such as 'what happened', 'why did it drop', and 'difference analysis', making accidental invocation likely. Because the skill has env:read and network:outbound permissions and performs external API calls, over-triggering can unnecessarily expose user queries and cause unintended use of sensitive credentials or network actions.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding
Suspicious Unicode normalization or mixed-script content

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill requires reading an API key from the environment and sending queries over the network, but it does not clearly warn users that their prompts, metric names, dimensions, and possibly business context may be transmitted externally. This lack of transparency increases privacy and compliance risk, especially in a skill designed for business analytics where identifiers and operational data may be sensitive.

Intent-Code Divergence

Medium
Confidence
93% confidence
Finding
The documentation states that all API usage reuses the Gateway API system, but later directs the agent to perform WebSearch for external events. This inconsistency undermines the trust boundary described to the user and reviewer, making it easier for data to leave the declared single-domain integration path without clear consent or review.

External Transmission

Medium
Category
Data Exfiltration
Content
> 示例：
> ```bash
> curl -H "X-API-Key: $CAN_API_KEY" "https://gateway.can.aloudata.com/api/metrics/search?pageSize=5" --data-urlencode "keyword=销售额" -G
> curl -H "X-API-Key: $CAN_API_KEY" "https://gateway.can.aloudata.com/api/metrics/retail_amt/dimensions"
> # 维度多关键词过滤（逗号分隔，匹配任一即返回，匹配范围包括维度名/展示名/描述/维度值样本）
> curl -H "X-API-Key: $CAN_API_KEY" "https://gateway.can.aloudata.com/api/metrics/retail_amt/dimensions" --data-urlencode "keyword=渠道,地区,品牌" -G
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
```bash
# 搜索指标
curl -H "X-API-Key: $CAN_API_KEY" "https://gateway.can.aloudata.com/api/metrics/search?pageSize=5" --data-urlencode "keyword=销售额" -G

# 查整体环比变化
curl -X POST "https://gateway.can.aloudata.com/api/metrics/query" \
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The skill instructs the agent to use a WebSearch capability for external event retrieval, but that capability is not declared in the skill metadata or permissions. This creates a tool/permission mismatch that can cause unauthorized external data access assumptions, inconsistent execution behavior, and unreviewed transmission of user context to third-party services.

Static analysis

No suspicious patterns detected.