Back to skill

Security audit

股份支付登记(权益结算/现金结算/等待期摊销)

Security checks for vulnerabilities and agentic risk

Overview

The verdict is suspicious because scanners report behavior that does not match the stated share-based payment review purpose, alongside concerns about document handling and incomplete checks.

Consider avoiding installation until the network integration and external document processing are clearly explained, and until the reconciliation checker distinguishes incomplete checks from successful ones. Skillspector's scan was partial: it covered 75% of components, with one file partly inspected and a degraded tool-misuse analyzer, so the results cannot establish that the artifact is safe.

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 (6)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

代码执行的是 MCP 服务目录发现和带凭据的工具枚举,没有读取年报附注、抽取股份支付数据、生成原文锚点或进行费用与数量校验。所访问的资源及认证工具发现行为未在声明中说明,代码中也未体现它们与股份支付登记任务的支持关系,因此构成实质性不匹配。

Content

No source excerpt is available for this finding.

The screenshot/scanned-document workflow sends user documents to an external reader without a privacy warning.

Medium
Category
Output Handling
Confidence
97% confidence
Finding

The workflow directs screenshots and scanned documents to an external reader without requiring confirmation that the user authorizes third-party processing. Annual reports are usually public, which reduces typical exposure, but user-supplied documents or screenshots can contain confidential annotations or unrelated sensitive material.

Content

No source excerpt is available for this finding.

The documented agent setup makes external network requests despite the README's explicit “zero network” claim.

Medium
Category
Excessive Agency
Confidence
90% confidence
Finding

L008 describes the skill as “零网络” (zero network), but the agent integration commands fetch an external catalog and send an authenticated tools/list request using CUE_API_KEY. These are documented setup commands rather than evidence of automatic execution, but following the prescribed workflow crosses a network and credential boundary that the unconditional zero-network claim denies.

Content

No source excerpt is available for this finding.

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 · README.md (reported line 107)May include surrounding context.

bash
  # ① 匿名取活体清单
  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)"

  # ② 带凭据握手取工具表

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.

Missing-data declarations silently bypass arithmetic checks despite the documented fail-closed guarantee.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

When an arithmetic operand is missing, any nonempty missing-data declaration can make the check return no findings, even if that declaration concerns an unrelated field; the quantity check uses the same mechanism. A crafted draft can therefore omit required values and still receive exit code 0, allowing downstream consumers to mistake an unperformed reconciliation for a successful one.

Content

No source excerpt is available for this finding.