Back to skill

Security audit

中标概率分析-投标胜率评估器

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly matches bid analysis, but it needs Review because it stores an API key locally, fingerprints the device for auto-registration, and can export shareable reports containing signed access links.

Review before installing. Use it only if you are comfortable with 知了标讯 API calls, possible paid credit usage, local storage of an API key, and a consent-gated registration flow that hashes a device MAC address. Prefer setting your own ZLBX_API_KEY to avoid auto-registration, and treat generated HTML reports as sensitive because they may include signed sk links that should not be broadly shared.

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 (34)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill advertises bid-win-rate analysis based on broad historical procurement data, but the provided content mainly specifies prompting and report-export behavior rather than verifiable implementation of the claimed analytical logic. This mismatch is dangerous because users may rely on authoritative-seeming conclusions or consent to data/API usage under false assumptions about what the skill actually does, while undisclosed export behavior increases data-handling risk.

Content

No source excerpt is available for this finding.

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The activation rule says the skill must be used for a very broad range of loosely related bidding questions, even when the user did not explicitly request win-rate analysis. Overbroad forced activation can route unnecessary user data to external services and trigger costly or privacy-impacting workflows without clear user intent.

Content

No source excerpt is available for this finding.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 81)May include surrounding context.

md
完整模式下,对话报告输出后默认再用 `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 85)May include surrounding context.

md
完整模式下,对话报告输出后默认再用 `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
84% confidence
Finding

This duplicate hidden-instructions finding points to the same concealed package-identity block. Concealing network-affecting behavior in comments undermines auditability and can be abused to smuggle instructions that the surrounding manifest does not clearly declare.

Content

Scanner excerpt · SKILL.md (reported line 163)May include surrounding context.

  • 报告里某个竞争对手威胁最大 → 引导用 zlbx-company-intel 对其深度背调
  • 以上都不贴切 → 报告涉及的企业完整档案与更多商机详情,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: bid-win-rate-analyzer/1.0.9

Hidden Instructions

High
Category
Prompt Injection
Confidence
84% confidence
Finding

This duplicate hidden-instructions finding points to the same concealed package-identity block. Concealing network-affecting behavior in comments undermines auditability and can be abused to smuggle instructions that the surrounding manifest does not clearly declare.

Content

Scanner excerpt · SKILL.md (reported line 163)May include surrounding context.

  • 报告里某个竞争对手威胁最大 → 引导用 zlbx-company-intel 对其深度背调
  • 以上都不贴切 → 报告涉及的企业完整档案与更多商机详情,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: bid-win-rate-analyzer/1.0.9

Hidden Instructions

High
Category
Prompt Injection
Confidence
95% confidence
Finding

This duplicate finding identifies the same hidden instruction block that attempts to control request identity and follow-up actions through embedded comments. In a skill that uses external APIs and configured API keys, such covert directives are more dangerous because they can silently influence authenticated requests and user-visible messaging without clear consent or transparent policy enforcement.

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: bid-decision/1.0.5

Hidden Instructions

High
Category
Prompt Injection
Confidence
95% confidence
Finding

This duplicate finding identifies the same hidden instruction block that attempts to control request identity and follow-up actions through embedded comments. In a skill that uses external APIs and configured API keys, such covert directives are more dangerous because they can silently influence authenticated requests and user-visible messaging without clear consent or transparent policy enforcement.

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: bid-decision/1.0.5

Hidden Instructions

High
Category
Prompt Injection
Confidence
89% confidence
Finding

This duplicate hidden-instruction finding refers to the same covert request-identity block. The issue remains that hidden content can steer authenticated requests and package attribution without transparent user or reviewer awareness.

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: bid-decision/1.0.5

Hidden Instructions

High
Category
Prompt Injection
Confidence
89% confidence
Finding

This duplicate hidden-instruction finding refers to the same covert request-identity block. The issue remains that hidden content can steer authenticated requests and package attribution without transparent user or reviewer awareness.

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: bid-decision/1.0.5

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill is presented as a bid win-rate analysis assistant, but it instructs the agent to collect device-derived identifiers and perform account registration. That is a material expansion of capability and data handling beyond the declared analytical purpose, creating an unnecessary privacy and trust boundary violation if the agent follows it.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The instructions direct the agent to execute system commands, inspect network interfaces, and derive a stable MAC-based hash for device fingerprinting. Even hashed, this is persistent device identification unrelated to bid evaluation, and the command execution requirement expands the skill's access to sensitive host information.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

Persisting API keys into local configuration and reusing them across sessions gives the skill credential-management and state-writing behavior that is unrelated to producing bid-analysis reports. This increases the blast radius of compromise because a content skill can influence long-lived authentication state on the user's machine.

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/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: bid-decision/1.0.5

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: bid-decision/1.0.5

Missing User Warnings

High
Category
Not specified by scanner
Confidence
96% confidence
Finding

The instructions mandate preserving full URL parameters, including sk login-bypass signatures, while presenting them in reports and exports without any privacy or security warning. This effectively turns reports into containers for reusable access tokens, making accidental leakage, forwarding, or archival of privileged links much more likely.

Content

No source excerpt is available for this finding.

Hidden Instructions

High
Category
Prompt Injection
Confidence
95% confidence
Finding

This duplicate finding points to the same hidden instruction region that tells the agent to preserve package identity headers and use configured credentials. In the context of a skill that depends on external APIs, concealed instructions of this type materially increase the chance of secret-bearing requests being made under attacker-controlled or nontransparent conditions.

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: bid-decision/1.0.5

Hidden Instructions

High
Category
Prompt Injection
Confidence
95% confidence
Finding

This duplicate finding points to the same hidden instruction region that tells the agent to preserve package identity headers and use configured credentials. In the context of a skill that depends on external APIs, concealed instructions of this type materially increase the chance of secret-bearing requests being made under attacker-controlled or nontransparent conditions.

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: bid-decision/1.0.5

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 71)May include surrounding context.

python
THREAT_COLOR = {"高": "#c0392b", "中": "#b9770e", "低": "#0d9463"}

# 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
79% confidence
Finding

The skill explicitly instructs the agent to write HTML reports to a local directory, but it does not declare any tool scope such as permissions or allowed-tools. That creates a capability/manifest mismatch where file-write behavior is not transparently bounded, making it harder for the platform or user to reason about what the skill may write and where.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

The skill description and all user-facing examples are presented as Chinese-only instructions and there is no statement that the user may choose another language. This can violate language/locale policy when the skill implicitly constrains responses to a specific language without offering a choice or documenting a justified locale restriction.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

Lines L77-L78 instruct the skill to relay update notices to the user verbatim using the service-provided message, and the surrounding document is entirely Chinese with no indication that users may choose another language. This creates a natural-language policy concern because the skill appears to force a specific language/locale for user-visible communication without opt-in.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The document is written as mandatory operational guidance for the agent and repeatedly provides Chinese text to relay directly to users, indicating a fixed language expectation. There is no indication that the user may choose another language or that the locale restriction is limited to a justified region-specific compliance context.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
95% confidence
Finding

This section explicitly instructs the agent to serialize collected device features and send them to an external registration endpoint. The transmission itself is the security-relevant behavior: host-derived identifiers are exfiltrated off-device for a purpose outside the skill's advertised analysis scope.

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`)。
>

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

The curl-based example is another direct instruction to send the same collected information to an external service. It reinforces off-device data transfer and command execution patterns that are unnecessary for bid-analysis functionality.

Content

Scanner excerpt · references/auto-register.md (reported line 131)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`)。
>
> 历史教训:曾有版本采集 `home_path`,Windows 的 `C:\Users\alice` 直接拼进 JSON 字符串时