Back to skill

Security audit

业绩预告速读·区间与修正逐条挂源

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because it reports behavior that conflicts with its stated earnings-forecast purpose, including external service discovery despite a declared zero-network workflow; analysis was partial.

Consider avoiding installation until the undeclared network and tool-discovery behavior and the revision-status issue are reviewed. The scan covered 13 of 14 components, with one partially inspected, so the evidence is incomplete; suspicious is a review verdict and does not establish malicious intent.

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 (12)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied commands perform remote service and tool discovery rather than the declared earnings-forecast extraction or local validation. These network operations directly contradict the explicit zero-network claim and access resources outside the declared workflow. The chunk does not show earnings-data processing, execution of discovered tools, or data exfiltration.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Both curl commands perform network operations, directly contradicting the declared zero-network constraint. The supplied chunk discovers remote services and tools rather than extracting or validating earnings-forecast fields. Although tool discovery could support a broader workflow, its external network access is a material mismatch with the explicit description.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

代码直接执行网络请求,与“零网络”的明确承诺相冲突。所示代码的行为是远程目录发现和工具枚举,而非预告信息登记或本地校验。ROUTING_URL 的具体地址及其赋值过程未出现在片段中,但外部目录请求已足以确认不匹配;不能据此推断代码还执行了其他远程工具或传输了财务数据。

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

该代码块的实际行为是外部服务与工具发现,而非业绩预告的本地整理和校验。对 https://cuecue.cn/api/mcp-catalog 的请求明确产生网络访问;后续命令还尝试向变量指定的路由地址发送 tools/list 请求。即使将工具发现视为辅助实现,其网络行为仍直接违反描述中的“零网络”约束。

Content

No source excerpt is available for this finding.

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 134)May include surrounding context.

md
bash scripts/fixtures/run_fixtures.sh # 三档离线自检,30 秒,零网络零计费

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.en.md (reported line 65)May include surrounding context.

CATALOG="https://cuecue.cn/api/mcp-catalog" curl -s "$CATALOG" | head -c 2000

take the group's routing address from the catalog, confirm the tools are present

curl -s -X POST "$ROUTING_URL"
-H 'Content-Type: application/json'
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

text

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 65)May include surrounding context.

CATALOG="https://cuecue.cn/api/mcp-catalog" curl -s "$CATALOG" | head -c 2000

从目录里取本组的路由地址,确认工具在场

curl -s -X POST "$ROUTING_URL"
-H 'Content-Type: application/json'
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

text

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 · SKILL.en.md (reported line 90)May include surrounding context.

CATALOG="https://cuecue.cn/api/mcp-catalog" curl -s "$CATALOG" | head -c 2000

take the group's routing address from the catalog, then confirm the tools are present

curl -s -X POST "$ROUTING_URL"
-H 'Content-Type: application/json'
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

text

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 · SKILL.md (reported line 105)May include surrounding context.

CATALOG="https://cuecue.cn/api/mcp-catalog" curl -s "$CATALOG" | head -c 2000

从目录里取本组的路由地址,再确认工具在场

curl -s -X POST "$ROUTING_URL"
-H 'Content-Type: application/json'
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

text

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 revision status is converted into a sourced assertion that no revision occurred.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The truthiness check treats an absent or null revised field as proof that no revision occurred, then attaches [S1] to that unsupported assertion. An incomplete or manipulated input can therefore conceal revision history behind apparent source attribution, undermining this financial skill's central provenance guarantee; draft warnings reduce but do not eliminate that risk.

Content

No source excerpt is available for this finding.

The output contract mandates Chinese-language headers and section titles without offering a language choice.

Low
Category
Output Handling
Confidence
92% confidence
Finding

The template requires Chinese-language output labels, with header text explicitly required verbatim, regardless of the user's language. Although the skill covers A-share disclosures, it does not explain why the surrounding report structure must be in Chinese rather than the user's preferred language.

Content

No source excerpt is available for this finding.