Back to skill

Security audit

A 股大师 Pro

Security checks for vulnerabilities and agentic risk

Overview

The stock-monitoring purpose is coherent, but the skill needs Review because it uses unsafe installation, credential, scheduled-task, and command-execution patterns.

Install only after the publisher removes the exposed API key, replaces raw curl execution with pinned and verified releases, fixes shell command construction and dashboard escaping, and makes scheduled monitoring explicitly opt-in with clear disable instructions. Treat holdings data and API keys as sensitive financial information.

Vulnerability Patterns
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • System PersistenceInstalls backdoors, hooks, services, or scheduled tasks that survive the run
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (5)

T03 · Remote Payload Retrieval and Execution

Error
Location
SKILL.md:55
Finding

Unpinned Remote JavaScript Is Downloaded and Executed

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/check_holdings.mjs:34
Finding

Shell Command Injection Through Holdings Stock Codes

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
INSTALL_SUMMARY.md:46
Finding

Published API Credential and Plaintext Persistent Secret Storage

Content
View full analysis
> ~/.bashrc source ~/.bashrc ``` The cron guidance also embeds credentials directly in scheduled command text: ```bash */10 * 9-15 * * 1-5 export QVERIS_API_KEY="sk-your API Key" && node ~/.openclaw/workspace/skills/stock-master-pro/scripts/check_holdings.mjs >> /tmp/stock_check.log 2>&1 ``` ### Technical Analysis The installation summary includes a concrete value with the expected `sk-` credential format rather than an obvious placeholder. The same document instructs users to write the value into a shell startup file. Even if the published value was intended only for testing, it must be treated as compromised because it is distributed with the project. The scheduled-task examples encourage users to place credentials directly in crontab command lines. Shell startup files and crontabs are persistent plaintext locations that may be exposed through backups, diagnostic bundles, configuration synchronization, local account access, or accidental publication. The project documentation elsewhere claims that there is no hard-coded API key, which conflicts with the actual content and can cause maintainers to overlook the exposure. ### Attack Path 1. An attacker obtains the distributed project or accesses its public source. 2. The attacker extracts the published `sk-...` value from `INSTALL_SUMMARY.md`. 3. If the credential remains active, the attacker uses it to access or consume the associated QVeris service. 4. Separately ...[truncated 661 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
web/dashboard.js:230
Finding

Stored DOM Cross-Site Scripting in Dashboard Rendering

Content
View full analysis
{ const profitClass = h.profit.profit >= 0 ? 'text-up' : 'text-down'; const changeClass = h.quote.changePct >= 0 ? 'text-up' : 'text-down'; return `
${h.name}
${h.code}
``` Alert messages are also rendered as HTML: ```js container.innerHTML = alerts.map(alert => { return `
${icon}
${alert.message}
${new Date(alert.timestamp).toLocaleString('zh-CN')}
`; }).join(''); ``` Remote announcement fields reach another `innerHTML` sink: ```js html += `
${icon}
${item.name} ${item.code}
Remediation
View remediation

T06 · System Persistence

Warning
Location
_meta.json:51
Finding

Scheduled Cross-Session Execution Is Ambiguous and Uses Unsafe Credential Handling

Content
View full analysis
> /tmp/stock_check.log 2>&1 30 12 * * 1-5 export QVERIS_API_KEY="sk-your API Key" && node ~/.openclaw/workspace/skills/stock-master-pro/scripts/market_review.mjs --session=noon >> /tmp/stock_review.log 2>&1 30 15 * * 1-5 export QVERIS_API_KEY="sk-your API Key" && node ~/.openclaw/workspace/skills/stock-master-pro/scripts/market_review.mjs --session=close >> /tmp/stock_review.log 2>&1 ``` ### Technical Analysis Scheduled execution is functionally related to the declared automatic stock-monitoring feature, so persistence is not inherently unrelated. However, the project ...[truncated 1851 chars]
Remediation
View remediation
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Rogue AgentSelf-Modification, Session Persistence
Findings (83)

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The document contains and propagates a concrete API key value directly in installation instructions. Publishing a real-looking secret in a skill artifact can expose paid service access, enable unauthorized API usage, and train users to handle credentials insecurely; in a stock-monitoring skill, this credential disclosure is unrelated to core functionality and therefore more suspicious, not less.

Content

No source excerpt is available for this finding.

YARA rule 'backdoor_persistence': Backdoor persistence with malicious payloads (shell commands, SSH key injection, hidden root users) [malware]

High
Category
YARA Match
Confidence
75% confidence
Finding

YARA rule matched a known malware signature (reverse shell, backdoor, ransomware, C2 framework, or info stealer).

Content

Scanner excerpt · INSTALL_SUMMARY.md (reported line 53)May include surrounding context.

控 | ⏸️ 待实现 | 监控持仓股公告 | | 龙虎榜分析 | ⏸️ 待实现 | 分析机构动向 | | 财报日历 | ⏸️ 待实现 | 财报发布提醒 |


🚀 立即开始使用

步骤 1:设置 API Key

bash
# 临时设置(当前会话)
export QVERIS_API_KEY="sk-Gc4yItJDkM3vBSJhnNPM_Ri4d1F6mar1YxdzK8RxOZI"

# 或永久设置(添加到 ~/.bashrc 或 ~/.zshrc)
echo 'export QVERIS_API_KEY="sk-Gc4yItJDkM3vBSJhnNPM_Ri4d1F6mar1YxdzK8RxOZI"' >> ~/.bashrc
source ~/.bashrc

步骤 2:检查依赖

bash
node ~/.openclaw/workspace/skills/stock-master-pro/scripts/check_dependency.mjs

应该显示:

text
✅ QVeris AI 依赖已就绪
✨ 所有依赖已就绪,可以开始使用 Stock Master Pro!

步骤 3:配置持仓

bash
# 复制配置模板
cp ~/.openclaw/workspace/skills/stock-master-pro/stocks/holdings.example.json \
   ~/.openclaw/workspace/skills/stock-master-pro/stocks/holdings.json

# 编辑持仓配置(已包含�

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The skill advertises automated monitoring, alerts, reviews, and screening, but the provided file contains only instructions, examples, and references to scripts without showing the promised logic. This mismatch is dangerous because users may grant trust, install dependencies, or provide portfolio data under false assumptions about what the skill actually does.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The skill advertises automated monitoring, alerts, reviews, and screening, but the provided file contains only instructions, examples, and references to scripts without showing the promised logic. This mismatch is dangerous because users may grant trust, install dependencies, or provide portfolio data under false assumptions about what the skill actually does.

Content

No source excerpt is available for this finding.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · package.sh (reported line 26)May include surrounding context.

sh
# 删除不需要打包的文件
echo "🗑️ 清理临时文件..."
rm -rf "$BUILD_DIR/build"
rm -rf "$BUILD_DIR/stocks"  # 用户数据不打包
rm -f "$BUILD_DIR/web/stocks"  # 符号链接不打包
rm -rf "$BUILD_DIR/.git"

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · package.sh (reported line 27)May include surrounding context.

sh
# 删除不需要打包的文件
echo "🗑️ 清理临时文件..."
rm -rf "$BUILD_DIR/build"
rm -rf "$BUILD_DIR/stocks"  # 用户数据不打包
rm -f "$BUILD_DIR/web/stocks"  # 符号链接不打包
rm -rf "$BUILD_DIR/.git"
rm -f "$BUILD_DIR/*.log"

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · package.sh (reported line 28)May include surrounding context.

sh
echo "🗑️ 清理临时文件..."
rm -rf "$BUILD_DIR/build"
rm -rf "$BUILD_DIR/stocks"  # 用户数据不打包
rm -f "$BUILD_DIR/web/stocks"  # 符号链接不打包
rm -rf "$BUILD_DIR/.git"
rm -f "$BUILD_DIR/*.log"
rm -f "$BUILD_DIR/web/test*.html"

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · package.sh (reported line 29)May include surrounding context.

sh
rm -rf "$BUILD_DIR/build"
rm -rf "$BUILD_DIR/stocks"  # 用户数据不打包
rm -f "$BUILD_DIR/web/stocks"  # 符号链接不打包
rm -rf "$BUILD_DIR/.git"
rm -f "$BUILD_DIR/*.log"
rm -f "$BUILD_DIR/web/test*.html"
rm -f "$BUILD_DIR/web/diag.html"

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · package.sh (reported line 30)May include surrounding context.

sh
rm -rf "$BUILD_DIR/stocks"  # 用户数据不打包
rm -f "$BUILD_DIR/web/stocks"  # 符号链接不打包
rm -rf "$BUILD_DIR/.git"
rm -f "$BUILD_DIR/*.log"
rm -f "$BUILD_DIR/web/test*.html"
rm -f "$BUILD_DIR/web/diag.html"

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · package.sh (reported line 31)May include surrounding context.

sh
rm -f "$BUILD_DIR/web/stocks"  # 符号链接不打包
rm -rf "$BUILD_DIR/.git"
rm -f "$BUILD_DIR/*.log"
rm -f "$BUILD_DIR/web/test*.html"
rm -f "$BUILD_DIR/web/diag.html"

# 创建示例持仓文件

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · package.sh (reported line 32)May include surrounding context.

sh
rm -rf "$BUILD_DIR/.git"
rm -f "$BUILD_DIR/*.log"
rm -f "$BUILD_DIR/web/test*.html"
rm -f "$BUILD_DIR/web/diag.html"

# 创建示例持仓文件
echo "📝 创建示例配置..."

Lp1

High
Category
MCP Least Privilege
Confidence
75% confidence
Finding

The skill uses 'env' capability that is not listed in its permissions. This may indicate deceptive intent or missing permission declarations.

Content

No source excerpt is available for this finding.

Lp1

High
Category
MCP Least Privilege
Confidence
75% confidence
Finding

The skill uses 'network' capability that is not listed in its permissions. This may indicate deceptive intent or missing permission declarations.

Content

No source excerpt is available for this finding.

Hidden Instructions

High
Category
Prompt Injection
Confidence
70% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · web/index.html (reported line 8)May include surrounding context.

html
<meta name="viewport" content="width=device-width, initial-scale=1.0">
  <title>Stock Master Pro - 智能盯盘系统</title>
  
  <!-- Tailwind CSS -->
  <script src="https://cdn.tailwindcss.com"></script>
  
  <!-- Chart.js -->

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

This markdown file states that the skill uses real-time data APIs, reads from local cache, runs crontab scheduled tasks, and automatically generates reports. Those behaviors can affect privacy, local data, or system operation, but the changelog contains no warning or disclosure about these impacts for users.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The installation guide states that Crontab tasks are auto-enabled after installation, but it does not clearly warn users before installation that persistent scheduled tasks will be created and run regularly. This can surprise users, create unauthorized persistence, and repeatedly access local data or external services without explicit informed consent, which is a meaningful security and privacy concern for an agent skill.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The instructions tell users to persist an API key in shell startup files without any warning about plaintext storage, local exposure, shell history, or multi-user system risk. This encourages insecure credential management and can cause long-lived secret compromise if the home directory, dotfiles, backups, or support bundles are later exposed.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill description is presented entirely in Chinese from the outset, which effectively imposes a specific language on users. The file does not state that the skill is China-region-specific as a documented locale constraint, nor does it offer an opt-in or alternative language.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill explicitly handles holdings and portfolio monitoring, but the release/marketing content does not warn users about the sensitivity of portfolio data or recommend safe handling practices. In a finance-focused skill, this omission can lead users to paste or upload sensitive position information without understanding privacy risks, increasing the chance of inadvertent disclosure through logs, dashboards, screenshots, or third-party integrations.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SETUP.md (reported line 21)May include surrounding context.

bash
# 创建目录
mkdir -p ~/.openclaw/skills/qveris-official/scripts

# 下载技能文件
curl -fSL https://qveris.ai/skill/SKILL.md -o ~/.openclaw/skills/qveris-official/SKILL.md

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The setup instructs users to download executable skill files directly from a remote domain and then run them, but provides no integrity verification, pinning, or trust warning. If the remote server, DNS, TLS endpoint, or distribution path is compromised, users may install attacker-controlled code with the same permissions as the user.

Content

No source excerpt is available for this finding.

File System Enumeration

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Code scans file system directories looking for sensitive files. This could be reconnaissance for credential theft.

Content

Scanner excerpt · SETUP.md (reported line 30)May include surrounding context.

curl -fSL https://qveris.ai/skill/scripts/qveris_env.mjs -o ~/.openclaw/skills/qveris-official/scripts/qveris_env.mjs

验证安装

ls -la ~/.openclaw/skills/qveris-official/

text

#### 步骤 3:设置 API Key

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The guide tells users to print the API key with echo $QVERIS_API_KEY and to embed the secret directly in cron entries. This increases the chance of credential disclosure through shell history, process listings, logs, screenshots, or readable crontab contents, especially on shared systems.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

L005 将“帮我选股”等自然口语列为触发语,并使用“等”扩展未限定的其他表达,缺少边界说明。这类短语在普通投资交流中也很常见,文档没有给出排除条件或更精确的触发范围,可能导致技能被意外调用。

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

L008-L013 使用“用户想要监控股票持仓”“询问持仓股票怎么样”“想要查看图形化盯盘界面”等宽泛条件描述技能适用场景,但未说明哪些相似表述不应触发。这会让技能在一般股票讨论或泛化问答场景中被过度匹配。

Content

No source excerpt is available for this finding.