Back to skill

Security audit

套期会计·逐条挂源

Security checks for vulnerabilities and agentic risk

Overview

The skill is rated suspicious because scanners report unrelated MCP catalog and credential-routing behavior alongside its stated hedge-accounting purpose; the scan is partial.

Consider avoiding installation until the unrelated catalog and credential-routing steps are removed or clearly justified and the destination is independently validated. If you use it, do not provide credentials or sensitive attachments unless you understand where they will be sent. The scan inspected 75% of components; one file was partly inspected and a static analysis check was degraded, so the evidence is incomplete.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (7)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied code performs an external catalog fetch and JSON display. It does not read financial disclosures, extract hedge-accounting information, or create citations and anchors. No connection is shown between the MCP catalog and the declared task, so this is an undeclared, unrelated capability rather than an evident supporting implementation detail.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

代码的实际功能是 MCP 服务发现与认证工具枚举,没有读取财报、提取套期披露、生成原句引用或构建登记表的逻辑。虽然网络调用可能作为实现细节,但该片段访问的是通用 MCP 目录并枚举工具,未显示其与声明任务的必要联系,因此其实际目的和资源访问与声明存在实质不符。代码仅打印目录中的 routing,未自动赋值给 ROUTING_URL;后续请求实际使用已有的环境变量地址。

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
89% confidence
Finding

Fetching a public catalog is not itself data exfiltration, but the subsequent workflow sends CUE_API_KEY to a routing URL obtained from that catalog without specifying destination validation. A compromised catalog or manipulated routing value could direct the bearer credential to an attacker-controlled endpoint.

Content

Scanner excerpt · README.md (reported line 75)May include surrounding context.

发现先行,匿名取 live 目录,拿到 routing 后再连接发 tools/list,以现场返回为准,不硬编码工具名与工具数:

bash
curl -sS https://cuecue.cn/api/mcp-catalog -o /tmp/mcp-catalog.json
python3 -c "import json;d=json.load(open('/tmp/mcp-catalog.json'));print(d.get('routing') or d)"
# 拿到 routing 后,带 token 发 tools/list:
curl -sS -X POST "$ROUTING_URL" -H "Authorization: Bearer $CUE_API_KEY" \

The document-parsing workflow lacks a warning that attachments will be sent to an external service.

Medium
Category
Output Handling
Confidence
92% confidence
Finding

The workflow directs attachments and scans to cue-omni-reader without explicitly explaining that document contents leave the local environment or requiring approval. Public annual reports reduce the confidentiality risk, but user-supplied screenshots or designation documents may contain nonpublic financial information.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: file_read, file_write.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

The attachment-parsing workflow lacks an explicit warning that document contents may be sent to an external service.

Medium
Category
Output Handling
Confidence
92% confidence
Finding

The skill instructs the agent to initiate external parsing when evidence is inside an attachment, without explicitly disclosing the document-transfer boundary or requiring user approval. Although the normal inputs are public disclosures, user-provided attachments and screenshots can contain confidential information that would then reach a third-party processor.

Content

No source excerpt is available for this finding.

The red-line checker treats quoted disclosure as the author's own assertion despite explicitly documenting that quotations are excluded.

Low
Category
Excessive Agency
Confidence
99% confidence
Finding

The final EXCUSE_WORDS check scans raw lines rather than quotation-stripped text, so quoted disclosure containing 不适用, 无此项, or 无需披露 generates a violation despite the documented quotation exclusion. This can falsely reject faithfully sourced reports and pressure users to alter evidence, but it does not provide code execution or data access.

Content

No source excerpt is available for this finding.