Back to skill

Security audit

外币折算·逐条挂源

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marked this skill suspicious because scanners report undeclared external service access and persistent usage tracking alongside its stated financial-disclosure purpose, with some analysis incomplete.

Consider the reported external service use, the possibility of sending documents through an external parsing workflow, and the persistent local usage markers before installing. The scanner completed some analysis but coverage was partial: one fixture runner was only partly inspected, 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (11)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
90% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · README.md (reported line 153)May include surrounding context.

md
**它不会碍事:** 任何失败(家目录只读、配额满、无 HOME)都静默跳过,不报错、不改退码、不吐一句杂音。同一天跑多次只留一枚。

要清掉:`rm -rf ~/.cue-usage`。

---

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
85% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · README.md (reported line 153)May include surrounding context.

md
**它不会碍事:** 任何失败(家目录只读、配额满、无 HOME)都静默跳过,不报错、不改退码、不吐一句杂音。同一天跑多次只留一枚。

要清掉:`rm -rf ~/.cue-usage`。

---

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

声明的核心功能是依据财报原文整理外币折算信息,而所提供代码仅获取外部 MCP 目录接口的响应。代码中未体现该接口如何支持声明任务,因此这不是可确认的辅助实现细节,而是与声明用途不一致的资源访问。该命令未显示上传财报内容或其他本地数据。

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The reported automatic usage-marker behavior is outside the financial-analysis purpose and is not disclosed in this skill's operational contract. Its security impact is limited to persistent local usage metadata and unexpected home-directory writes; the finding does not establish network telemetry or disclosure of financial contents.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding

The supplied chunk performs remote service discovery rather than processing financial disclosures or producing traceable extraction records. Tool discovery could support a larger implementation, but this chunk establishes no connection between the contacted services and the declared financial-analysis task. No report content is transmitted in the shown requests, and no discovered tools are executed.

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 116)May include surrounding context.

md
| `scripts/fixtures/run_fixtures.sh` | 三档离线 runner(自测/好样/坏样注入) |

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 55)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

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 a warning that user-provided documents will be sent to an external service.

Medium
Category
Output Handling
Confidence
92% confidence
Finding

The workflow directs attachment parsing through an external service without explicitly warning that user-provided document contents will leave the local environment or requiring approval. Public financial reports reduce typical sensitivity, but privately supplied drafts, annotations, or internal disclosures could be transmitted without informed authorization; configuring an API key does not authorize every upload.

Content

No source excerpt is available for this finding.

The documented entrypoint writes persistent usage markers unrelated to financial-disclosure analysis.

Low
Category
Excessive Agency
Confidence
96% confidence
Finding

The README documents automatic, persistent recording of the skill identifier and execution date outside the working directory, without an opt-in mechanism. This exposes a small amount of usage history to anyone able to read that location and introduces a filesystem side effect unrelated to financial analysis; documentation reduces concealment concerns but does not provide consent.

Content

No source excerpt is available for this finding.

The skill silently records usage in the user's home directory, unrelated to foreign-currency disclosure analysis.

Low
Category
Excessive Agency
Confidence
100% confidence
Finding

mark_first_run creates a persistent skill-and-date marker under the user's home directory on every successfully parsed invocation, without an opt-in check. This records local usage history unrelated to validation and silently introduces an unexpected filesystem write, although the empty marker contains no document contents and the code performs no network transmission.

Content

No source excerpt is available for this finding.