Back to skill

Security audit

AI招投标分析师-全网招中标智能分析

Security checks for vulnerabilities and agentic risk

Overview

The search functions fit the skill’s purpose, but its signup prompt understates the privacy impact of a persistent device identifier.

Review the device-identification disclosure before accepting automatic signup, or configure an existing API key to bypass it. Search terms and company names go to Zhiliaobiaoxun as part of normal queries, which can consume account credits. Protect the saved key and generated login links. This review verified the packaged instructions; it did not establish how long the provider retains device identifiers.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Rogue AgentSelf-Modification, Session Persistence
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (9)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The declared purpose concerns procurement and market-data analysis, but this code collects and hashes a local hardware identifier. Device fingerprinting is not disclosed or evidently necessary to that purpose. The supplied chunk does not transmit the identifier or perform registration, so those behaviors cannot be inferred from the filename.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The declared purpose concerns procurement and market-data analysis, not device identification. The supplied code instead accesses local network information and hashes a hardware identifier, with no demonstrated connection to the declared analytical functionality. The chunk does not show transmission, storage, or actual registration of the resulting identifier.

Content

No source excerpt is available for this finding.

The query workflow sends user-derived search criteria to an external service without a privacy disclosure.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The execution rules direct the agent to construct and run searches using the user's criteria against the external API identified at L019, with company-name submissions illustrated at L323–L328. Although device-data collection during registration explicitly requires consent, the skill provides no comparable disclosure that query keywords, target companies, and other criteria will be transmitted to Zhiliaobiaoxun, potentially revealing confidential procurement or competitive interests.

Content

No source excerpt is available for this finding.

Automatic company searches transmit user-provided company names to an external API without a privacy disclosure.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The workflow instructs the agent to send the user's company name to search_company and use the expanded company list in subsequent queries without interrupting the user. Unlike a passive API catalog entry, this is an instruction to transmit user-provided search information to a third-party service, but the file does not warn users about that transmission.

Content

No source excerpt is available for this finding.

Automatic registration collects and transmits a persistent hardware fingerprint, a capability not justified by the tender-analysis manifest.

Medium
Category
Excessive Agency
Confidence
90% confidence
Finding

The registration workflow transmits a deterministic, unsalted MAC-address hash while telling users that the collected features have no identity meaning and contain no personal information. Such a hash remains a stable pseudonymous identifier that can link activity and can be tested against candidate MAC addresses; the consent gate reduces unauthorized collection but does not correct the misleading privacy assurance.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
70% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

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
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

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

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 · references/auto-register.md (reported line 195)May include surrounding context.

text

注意事项:
- 目录不存在时先 `mkdir -p ~/.zlbx`
- 文件已存在时**合并而非覆盖**(保留用户可能的其他配置)
- `source: "auto"` 字段必须写入,**这是后续判断「是否输出自动登录链接」的关键依据**

The skill requires a remote update-notice acknowledgment unrelated to tender analysis.

Low
Category
Excessive Agency
Confidence
88% confidence
Finding

The instructions add a mandatory per-session upgrade-notice workflow and direct the agent to submit an authenticated POST confirming that a notice was displayed. Recording notice-display state on the provider's server is package-lifecycle tracking, not a requirement of the manifest's tender research and market-analysis functions.

Content

No source excerpt is available for this finding.