Back to skill

Security audit

CNKI Scholar

Security checks for vulnerabilities and agentic risk

Overview

This academic search skill is coherent and purpose-aligned, with expected third-party API lookups but no persistence, credential use, or hidden execution.

Install only if you are comfortable with scholarly search queries and DOI lookups being sent to OpenAlex or similar academic services. Do not include secrets, confidential project names, regulated personal data, or proprietary research terms in searches unless you intend to disclose them to that service.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • 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 (15)

External Script Fetching

High
Category
Supply Chain
Content
使用OpenAlex API搜索:

```bash
curl -s "https://api.openalex.org/works?search=YOUR_QUERY&per_page=10"
```

支持的筛选参数:
Confidence
90% confidence
Finding
Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

External Script Fetching

High
Category
Supply Chain
Content
使用OpenAlex API搜索:

```bash
curl -s "https://api.openalex.org/works?search=YOUR_QUERY&per_page=10"
```

支持的筛选参数:
Confidence
90% confidence
Finding
Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

External Script Fetching

High
Category
Supply Chain
Content
```bash
get_paper_by_doi() {
    local doi="$1"
    curl -s "https://api.openalex.org/works/doi:${doi}"
}
```
Confidence
90% confidence
Finding
Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The manifest description says to activate when the user needs '学术搜索、论文元数据提取、引用格式生成' or says phrases like '搜论文' and '查文献', which are broad requests that can overlap with many general research or writing tasks. It does not clearly bound when this skill should activate versus when a more general search, citation, or literature-review skill should not.

External Transmission

Medium
Category
Data Exfiltration
Content
使用OpenAlex API搜索:

```bash
curl -s "https://api.openalex.org/works?search=YOUR_QUERY&per_page=10"
```

支持的筛选参数:
Confidence
82% confidence
Finding
The skill instructs sending user-provided search queries to the external OpenAlex API. Even though this is expected for a search skill, it still transmits user input off-platform and could expose sensitive research topics, proprietary terms, or personal data if users include them in queries.

External Transmission

Medium
Category
Data Exfiltration
Content
local limit="${2:-10}"
    local year_filter="${3:-}"
    
    local url="https://api.openalex.org/works?search=${query}&per_page=${limit}"
    
    if [ -n "$year_filter" ]; then
        url="${url}&filter=publication_year:${year_filter}"
Confidence
88% confidence
Finding
The constructed URL embeds raw user-controlled query text directly into an external request. This creates a real data-exposure surface and can also cause malformed requests or parameter injection into the API query string because the input is not encoded before transmission.

External Transmission

Medium
Category
Data Exfiltration
Content
```bash
get_paper_by_doi() {
    local doi="$1"
    curl -s "https://api.openalex.org/works/doi:${doi}"
}
```
Confidence
84% confidence
Finding
The DOI provided by the user is sent to an external service, which is an outbound data transfer. This is usually necessary for DOI resolution, but it still leaks user interest and should be treated as external disclosure of input data.

External Transmission

Medium
Category
Data Exfiltration
Content
```bash
generate_apa() {
    local doi="$1"
    curl -s "https://api.openalex.org/works/doi:${doi}" | python3 -c "
import json, sys
w = json.load(sys.stdin)
authors = []
Confidence
86% confidence
Finding
The citation-generation workflow sends a user-supplied DOI to OpenAlex and processes returned metadata. The primary risk is external transmission of user-supplied identifiers and research intent, not local code execution.

External Transmission

Medium
Category
Data Exfiltration
Content
## 搜索接口

```
GET https://api.openalex.org/works?search=QUERY&per_page=N
```

### 常用参数
Confidence
50% 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
## 搜索接口

```
GET https://api.openalex.org/works?search=QUERY&per_page=N
```

### 常用参数
Confidence
50% 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
## 搜索接口

```
GET https://api.openalex.org/works?search=QUERY&per_page=N
```

### 常用参数
Confidence
50% 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
## 搜索接口

```
GET https://api.openalex.org/works?search=QUERY&per_page=N
```

### 常用参数
Confidence
50% 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
84% confidence
Finding
The skill name, description, and heading present the tool as Chinese-language only ('中文学术论文搜索工具', '中文学术论文搜索') while also claiming to search English papers, but there is no indication that users may choose output language or locale. This can violate language/locale policy when activated for users who did not opt into Chinese responses.

Natural-Language Policy Violations

Low
Confidence
89% confidence
Finding
The guidance table states '英文论文 | APA 7th' and '中文论文 | GB/T 7714' as fixed recommendations tied to language. This is a natural-language locale/language preference that does not explicitly offer user opt-in or indicate that other formats may be chosen based on venue requirements.

Natural-Language Policy Violations

Low
Confidence
81% confidence
Finding
The file title and all instructional content are presented only in Chinese, which constitutes a language/locale constraint in the skill documentation. The document does not offer an opt-in language choice or explain that the skill is specifically intended for a Chinese-speaking or China-specific context.

Static analysis

No suspicious patterns detected.