Back to skill

Security audit

招投标尽调助手-合同流水看经营实态

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent for bid due-diligence, but it handles device identifiers, persistent API credentials, contact phone data, and shareable signed links in ways users should review carefully.

Review before installing if you are uncomfortable with automatic account creation, hashed device identifier collection, local API-key storage, or reports that include signed access links. Prefer configuring your own ZLBX_API_KEY and avoid sharing generated HTML reports unless you are comfortable exposing their embedded links.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Findings (36)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 89)May include surrounding context.

md
默认再用 `scripts/render_report.py` 生成一份可分享的 HTML 版报告并告知保存路径(详见 report-template.md「HTML 报告导出」)。

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 93)May include surrounding context.

md
默认再用 `scripts/render_report.py` 生成一份可分享的 HTML 版报告并告知保存路径(详见 report-template.md「HTML 报告导出」)。

Hidden Instructions

High
Category
Prompt Injection
Confidence
92% confidence
Finding

This duplicate hidden-instructions finding points to the same concealed package-identity block. Concealed control instructions are dangerous because they can alter request semantics and trigger extra network interactions (for update display acknowledgments) that are easy to miss during review.

Content

Scanner excerpt · SKILL.md (reported line 185)May include surrounding context.

  • 用户想持续跟踪该公司 → 介绍竞对动态周报(见上方「竞对动态周报」)
  • 以上都不贴切 → 询问是否要对某个竞对做双公司对比,或引导访问知了商机大师 https://agent.zhiliaobiaoxun.com

安装包身份与升级检查:所有发往知了标讯 API 的请求(包括账户查询)必须额外携带以下两个请求头:

text
X-Skill-Package: bidding-due-diligence/1.0.6

Hidden Instructions

High
Category
Prompt Injection
Confidence
92% confidence
Finding

This duplicate hidden-instructions finding points to the same concealed package-identity block. Concealed control instructions are dangerous because they can alter request semantics and trigger extra network interactions (for update display acknowledgments) that are easy to miss during review.

Content

Scanner excerpt · SKILL.md (reported line 185)May include surrounding context.

  • 用户想持续跟踪该公司 → 介绍竞对动态周报(见上方「竞对动态周报」)
  • 以上都不贴切 → 询问是否要对某个竞对做双公司对比,或引导访问知了商机大师 https://agent.zhiliaobiaoxun.com

安装包身份与升级检查:所有发往知了标讯 API 的请求(包括账户查询)必须额外携带以下两个请求头:

text
X-Skill-Package: bidding-due-diligence/1.0.6

Hidden Instructions

High
Category
Prompt Injection
Confidence
94% confidence
Finding

This duplicated hidden-instructions finding identifies the same concealed request-identity directive. In context with the rest of the file, the concealment increases concern because it accompanies credential handling and registration behaviors that exceed the skill's declared purpose.

Content

Scanner excerpt · references/auto-register.md (reported line 1)May include surrounding context.

请求身份:使用已配置的 API Key,并保留本包的 X-Client,携带下列完整身份头;不得沿用其它包的版本或来源。

text
X-Client: company-intel/1.0.2

Hidden Instructions

High
Category
Prompt Injection
Confidence
94% confidence
Finding

This duplicated hidden-instructions finding identifies the same concealed request-identity directive. In context with the rest of the file, the concealment increases concern because it accompanies credential handling and registration behaviors that exceed the skill's declared purpose.

Content

Scanner excerpt · references/auto-register.md (reported line 1)May include surrounding context.

请求身份:使用已配置的 API Key,并保留本包的 X-Client,携带下列完整身份头;不得沿用其它包的版本或来源。

text
X-Client: company-intel/1.0.2

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The document instructs the agent to perform automatic account registration, collect device-derived identifiers, persist credentials, and manage quota/login flows that are unrelated to the declared bidding due-diligence purpose. This expands the skill's authority from data retrieval into user/account bootstrap and local state modification, increasing the risk of unauthorized data collection, secret handling, and persistence abuse.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill directs the agent to run OS-specific shell and PowerShell commands to collect device fingerprint material such as architecture and hashed MAC address. Even with claimed minimization, this is unnecessary for a due-diligence/reporting skill and creates privacy and execution risk by normalizing host inspection and external transmission of device identifiers.

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 · references/api-quick.md (reported line 1)May include surrounding context.

请求身份:使用已配置的 API Key,并保留本包的 X-Client,携带下列完整身份头;不得沿用其它包的版本或来源。

text
X-Client: company-intel/1.0.2

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 · references/report-template.md (reported line 1)May include surrounding context.

请求身份:使用已配置的 API Key,并保留本包的 X-Client,携带下列完整身份头;不得沿用其它包的版本或来源。

text
X-Client: company-intel/1.0.2

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 · references/api-quick.md (reported line 1)May include surrounding context.

请求身份:使用已配置的 API Key,并保留本包的 X-Client,携带下列完整身份头;不得沿用其它包的版本或来源。

text
X-Client: company-intel/1.0.2

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 · references/report-template.md (reported line 1)May include surrounding context.

请求身份:使用已配置的 API Key,并保留本包的 X-Client,携带下列完整身份头;不得沿用其它包的版本或来源。

text
X-Client: company-intel/1.0.2

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The template explicitly requires preserving full URLs containing sk login-bypass parameters in reports and exported HTML, but provides no warning that these links are access-bearing and sensitive. If such reports are forwarded or stored insecurely, recipients may gain unintended access through the embedded tokenized links.

Content

No source excerpt is available for this finding.

Ssd 3

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill instructs the agent to expose authenticated URLs with sk parameters directly in Markdown and HTML outputs. This is a direct secret-disclosure pattern because the generated artifacts become portable containers for bearer-like access tokens, enabling unauthorized viewing if shared, cached, or indexed.

Content

No source excerpt is available for this finding.

Hidden Instructions

High
Category
Prompt Injection
Confidence
84% confidence
Finding

This appears to be a duplicate detection of the same hidden-instructions issue at the top of the file. Even if intended for legitimate package routing, embedding credential and identity directives in hidden comments is still risky because the agent may obey non-transparent instructions to send authenticated requests or special headers without clear user awareness.

Content

Scanner excerpt · references/workflow.md (reported line 1)May include surrounding context.

请求身份:使用已配置的 API Key,并保留本包的 X-Client,携带下列完整身份头;不得沿用其它包的版本或来源。

text
X-Client: company-intel/1.0.2

Hidden Instructions

High
Category
Prompt Injection
Confidence
84% confidence
Finding

This appears to be a duplicate detection of the same hidden-instructions issue at the top of the file. Even if intended for legitimate package routing, embedding credential and identity directives in hidden comments is still risky because the agent may obey non-transparent instructions to send authenticated requests or special headers without clear user awareness.

Content

Scanner excerpt · references/workflow.md (reported line 1)May include surrounding context.

请求身份:使用已配置的 API Key,并保留本包的 X-Client,携带下列完整身份头;不得沿用其它包的版本或来源。

text
X-Client: company-intel/1.0.2

Obfuscated Code

High
Category
Supply Chain
Confidence
50% confidence
Finding

Code contains obfuscation (base64, hex encoding with execution). This is often used to hide malicious functionality.

Content

Scanner excerpt · scripts/render_report.py (reported line 93)May include surrounding context.

python
# 知了标讯白色 logo(299x96 PNG base64 内嵌,保证报告离线/打印/转发时不裂图)
_LOGO_B64 = "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
...[truncated 27 chars]

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
81% confidence
Finding

The skill explicitly writes reports to a local directory (~/zlbx-company-intel-files/) and references generating HTML output, but it does not declare a matching tool/file-write scope. This mismatch can bypass least-privilege expectations, making it harder for the host or user to understand and constrain the skill’s write capabilities.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The activation guidance says the skill must be used even when the user did not explicitly ask for due diligence, as long as they are considering investment or cooperation. Over-broad routing can cause unnecessary third-party data disclosure, unintended API usage/costs, and collection of company queries in contexts where the user may have expected a simple discussion instead of external lookups.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The natural-language instructions and required example phrasing are entirely in Chinese, and the skill repeatedly prescribes exact Chinese user-facing wording without indicating that other languages are supported or that Chinese is required by policy. Under the language/locale policy, forcing a specific language without user opt-in is a violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
75% confidence
Finding

The document is written entirely in Chinese and frames the skill behavior around a Chinese-language workflow, but it does not state that language choice is optional or that the locale restriction is intentional and justified. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly supports retrieving and displaying project contact phone numbers, including full numbers for paid accounts, even though the manifest frames the skill as bid due diligence and business verification. This expands the data scope into personal contact disclosure without a clear necessity test, increasing privacy, harassment, and misuse risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The documentation instructs the agent to transmit contact phone data 'as returned' and even encourages upgrade behavior when numbers are masked, but it does not require a user-facing privacy warning or explain the sensitivity of the data. That makes unauthorized disclosure more likely and normalizes exposing personal contact details in ordinary workflow output.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill provides a fixed Chinese consent prompt for the user and presents it as the required example to use before auto-registration. Because the file does not offer an alternative language or ask the user for locale preference, it imposes a specific language on user-facing interaction.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
92% confidence
Finding

This instruction supports sending collected device features to an external registration endpoint. External transmission alone is not always unsafe, but here the transmitted data includes device fingerprint material for a purpose unrelated to the skill's declared function, making the exfiltration unjustified and privacy-impacting.

Content

Scanner excerpt · references/auto-register.md (reported line 130)May include surrounding context.

md
> ### ⚠️ 请求体必须用 JSON 序列化函数生成,不要手拼字符串
>
> 用 `json.dumps(payload)` / `requests.post(url, json=payload)` / `JSON.stringify(payload)`,
> 或 `curl -d @file`;**不要用字符串拼接,也不要用 Python 的 `str(dict)`**
> (后者产出单引号,服务端会报 `Expecting property name enclosed in double quotes`)。
>