Back to skill

Security audit

临期续约商机-合同到期窗口挖掘

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly matches its business-search purpose, but it also handles account setup, persistent credentials, device-derived registration data, and shareable signed links in ways users should review carefully.

Install only if you are comfortable using the Zhiliaobiaoxun API service, spending account credits, storing an API key in ~/.zlbx/config.json, allowing consent-gated trial registration with a hashed MAC-derived identifier, and sharing reports or chat output that may contain signed sk access links. Avoid forwarding generated links or HTML reports outside trusted recipients.

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 (33)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

声明描述的是一个“发现/查询型”技能,预期应包含数据检索、筛选、排序、识别和商机扫描等核心业务逻辑。但代码实际只是一个报表展示层组件:从本地JSON读取结果并格式化成HTML,不接触外部数据源,不执行搜索,不计算到期窗口,不识别供应商,也不按行业/地区生成清单。虽然报告内容字段与声明场景相关(expiring/proposed/intent),但这些只是对上游结果的展示,不等于实现了声明中的主要能力。因此代码实际行为与技能声明的主要用途存在实质性不一致。

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,对话清单输出后默认再用 `scripts/render_report.py` 生成一份可分享的 HTML 版商机清单并告知保存路径(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,对话清单输出后默认再用 `scripts/render_report.py` 生成一份可分享的 HTML 版商机清单并告知保存路径(详见 report-template.md「HTML 报告导出」)。

Hidden Instructions

High
Category
Prompt Injection
Confidence
93% confidence
Finding

This duplicate finding points to the same hidden-instructions block controlling API request identity and acknowledgment flow. Because the directives are concealed in comments, they can bypass ordinary review expectations and create non-obvious outbound signaling or behavior constraints.

Content

Scanner excerpt · SKILL.md (reported line 171)May include surrounding context.

  • 用户想持续跟踪 → 介绍订阅模式(见上方「订阅模式」)
  • 以上都不贴切 → 询问是否要对某条高价值商机深挖,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: expiring-contract-renewal-finder/1.0.8

Hidden Instructions

High
Category
Prompt Injection
Confidence
93% confidence
Finding

This duplicate finding points to the same hidden-instructions block controlling API request identity and acknowledgment flow. Because the directives are concealed in comments, they can bypass ordinary review expectations and create non-obvious outbound signaling or behavior constraints.

Content

Scanner excerpt · SKILL.md (reported line 171)May include surrounding context.

  • 用户想持续跟踪 → 介绍订阅模式(见上方「订阅模式」)
  • 以上都不贴切 → 询问是否要对某条高价值商机深挖,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: expiring-contract-renewal-finder/1.0.8

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: opportunity-radar/1.0.3

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: opportunity-radar/1.0.3

Hidden Instructions

High
Category
Prompt Injection
Confidence
96% confidence
Finding

This duplicate hidden-instruction finding is valid for the same reason: request-shaping logic is embedded in an HTML comment that the agent is expected to obey. Concealed prompt directives create a covert control plane for outbound requests and increase the chance of unauthorized or unaudited metadata disclosure.

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: opportunity-radar/1.0.3

Hidden Instructions

High
Category
Prompt Injection
Confidence
96% confidence
Finding

This duplicate hidden-instruction finding is valid for the same reason: request-shaping logic is embedded in an HTML comment that the agent is expected to obey. Concealed prompt directives create a covert control plane for outbound requests and increase the chance of unauthorized or unaudited metadata disclosure.

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: opportunity-radar/1.0.3

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The document embeds a full auto-registration, credential acquisition, persistence, and recharge workflow inside a skill whose stated purpose is identifying expiring contracts and renewal opportunities. This materially expands the skill's capabilities into account management and local credential storage, creating unnecessary access, persistence, and data-handling risk outside the declared business function.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill instructs the agent to collect device fingerprinting material (platform, architecture, MAC-derived hash) and to generate recovery/recharge login links unrelated to contract-renewal discovery. Even with user-consent language, this is sensitive telemetry and account lifecycle handling that broadens surveillance and misuse potential if the skill is invoked in a normal business-search context.

Content

No source excerpt is available for this finding.

Hidden Instructions

High
Category
Prompt Injection
Confidence
95% confidence
Finding

This duplicate finding identifies the same hidden instruction block that tells the agent to send authenticated requests with package identity headers. Such hidden prompt content is a security risk because it embeds sensitive operational control in untrusted markdown, making it easier for adversarial content to influence credentialed actions outside explicit user intent.

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: opportunity-radar/1.0.3

Hidden Instructions

High
Category
Prompt Injection
Confidence
95% confidence
Finding

This duplicate finding identifies the same hidden instruction block that tells the agent to send authenticated requests with package identity headers. Such hidden prompt content is a security risk because it embeds sensitive operational control in untrusted markdown, making it easier for adversarial content to influence credentialed actions outside explicit user intent.

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: opportunity-radar/1.0.3

Hidden Instructions

High
Category
Prompt Injection
Confidence
95% confidence
Finding

This duplicate finding points to the same hidden instruction block that tells the agent to preserve and send package identity headers along with configured credentials. Even if intended for routing or attribution, placing such directives inside the skill text makes them adversarially mutable and can cause unauthorized request shaping, trust-boundary confusion, and accidental leakage of package/source identity across contexts.

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: opportunity-radar/1.0.3

Hidden Instructions

High
Category
Prompt Injection
Confidence
95% confidence
Finding

This duplicate finding points to the same hidden instruction block that tells the agent to preserve and send package identity headers along with configured credentials. Even if intended for routing or attribution, placing such directives inside the skill text makes them adversarially mutable and can cause unauthorized request shaping, trust-boundary confusion, and accidental leakage of package/source identity across contexts.

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: opportunity-radar/1.0.3

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 64)May include surrounding context.

python
ROUTE_COLOR = {"拟建": "#7c5cbf", "意向": "#0b7a6a", "临期": "#b9770e"}

# 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
88% confidence
Finding

The skill declares no explicit tool or permission scope, yet its instructions clearly include local file output to ~/zlbx-opportunity-radar-files/ and report generation. Without a restrictive allowlist, an agent/runtime may grant broader file-write capability than users expect, increasing the chance of unintended local data creation or overwrite if the skill is triggered inappropriately.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The activation text is extremely broad and says the skill must be used even when the user does not explicitly mention expiring contracts, as long as the request loosely relates to renewal or replacement opportunities. Overbroad triggering can cause unnecessary external API calls, account usage, and file generation in contexts where the user did not clearly consent to this specific workflow.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The natural-language description and required user-facing phrasing are written as mandatory Chinese interaction patterns, and the file does not offer an opt-in or alternative language choice. This can violate language/locale policy where users should not be forced into a specific language without consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The file explicitly says result URLs contain an sk login-bypass parameter and should be output verbatim for direct clicking. Exposing bearer-style access links to users or downstream systems can leak authenticated access via chat logs, browser history, referrers, or third-party sharing, effectively turning results into transferable tokens.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The documentation instructs the skill to call an unrelated account/balance endpoint and then submit a display acknowledgement tied to upgrade notices before or alongside normal business queries. This expands the skill’s behavior beyond contract-renewal discovery into account-state probing and state-changing side effects, creating unnecessary metadata disclosure and a channel for remotely driven prompts/messages to reach users.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The document mandates Chinese user-facing disclosures and later provides required Chinese message templates for user communication. Because the file does not offer a language choice or ask the user to opt into Chinese, it appears to impose a specific language/locale on all users, which matches the policy-violation category.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The document claims privacy minimization by not collecting home_path, but later directs reading and writing ~/.zlbx/config.json under the user's home directory. While this is not the same as transmitting home_path, the messaging is misleading because the workflow still accesses and persists data in the user's home directory, which weakens informed consent and trust.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
94% confidence
Finding

This instruction explicitly directs transmission of collected device features to an external service via JSON POST. Because the transmitted data includes device fingerprinting elements and is unrelated to the core contract-finder purpose, the external transmission increases privacy and account-abuse risk, especially if users do not expect outbound registration traffic from this skill.

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
94% confidence
Finding

The curl -d guidance is another explicit instruction to serialize and send registration payloads to a remote endpoint. The concern is not the serialization advice itself, but that the skill is facilitating external exfiltration of device-derived data and account bootstrap actions outside its declared scope.

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