Back to skill

Security audit

Travel Consultant

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this travel consultation skill suspicious, mainly because its documented install command uses an unpinned package and some executable files were only partly analyzed.

Consider not installing until the package version is pinned and independently verified, and until you understand what travel details and email are sent to the external service and when. The analysis covered only part of the artifact, so the available scan 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (20)

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 294)May include surrounding context.

md
| `node` (>=18, zero third-party deps) | Runs `scripts/travel_aipay.mjs` | Report clearly and stop |

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 408)May include surrounding context.

md
| `scripts/travel_aipay.mjs` | Request body assembly, 402 bill retrieval, official payment orchestration, retrieval (`complete --wait` polling + official payloa

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 411)May include surrounding context.

md
| `tests/mock_server.mjs` | Local mock resource server (fake bills, never touches the real service) |

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 412)May include surrounding context.

md
| `tests/run_offline_tests.mjs` | Offline regression: 402 / generating / success / failure / same-order retry / payment interception |

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/install-and-pack.md (reported line 108)May include surrounding context.

md
-x '*/node_modules/*' \
  -x '*.pyc' \
  -x '*/__pycache__/*' \
  -x '*/.env' \
  -x '*/payment-needed.txt' \
  -x '*/body.json' \
  -x '*/state.json' \

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

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.

MCP server referenced without pinned version: 'npx -y @alipay/agent-payment'.

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The documented installation command executes @alipay/agent-payment@latest rather than a reviewed, immutable version. A compromised future release could execute with the user's privileges and access payment-related state; user consent and an unspecified integrity check reduce risk but do not establish that the selected release is trusted.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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 301)May include surrounding context.

md
### Request contract

`POST https://api.riyuexinzhu.com/aipay/travel_consultant`, `Content-Type: application/json`; add `Payment-Proof` when retrieving the resource (it may only come from official CLI output).

| Field | Required | Description |
|---|---|---|

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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_cn.md (reported line 270)May include surrounding context.

md
### 请求契约

`POST https://api.riyuexinzhu.com/aipay/travel_consultant`,`Content-Type: application/json`;取货时额外带 `Payment-Proof`(只能来自官方 CLI 输出)。

| 字段 | 必填 | 说明 |
|---|---|---|

MCP server referenced without pinned version: 'npx -y @alipay/agent-payment'.

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The dependency-recovery instructions use @alipay/agent-payment@latest, allowing executable dependency contents to change without review. Compromise of a future release could expose local files and payment state; checking registry integrity alone would not distinguish a malicious published release from a trusted one.

Content

No source excerpt is available for this finding.

The workflow sends the user's travel request and email to an external service without requiring a user-facing privacy disclosure.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The probe step transmits the user's travel query and email to the external provider before the workflow requires explicit payment confirmation, without first requiring disclosure of that data transfer. Travel queries can include family details, travel dates, budgets, or accessibility needs; the service's legitimate need for these inputs does not eliminate the privacy risk from undisclosed sharing.

Content

No source excerpt is available for this finding.

The probe workflow sends the user's travel query and email to an external service without a privacy warning.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The run instructions invoke probe with --query and --email, and the file identifies an external resource endpoint at line 135. Although it requires consent before payment, it does not warn that obtaining the bill transmits these user details to the external travel service before payment.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · references/install-and-pack.md (reported line 67)May include surrounding context.

~/.agents/skills ~/.copilot/skills ~/.claude/skills

text

仓库根的 `skills/` **不在**默认清单里,不登记就**不会自动发现**(也不会有 `/` 斜杠命令),只能靠用户明确说"读取 `skills/xianzong-travel-aipay/SKILL.md`"手动加载。

本仓库已在 `.vscode/settings.json` 里登记:

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · references/install-and-pack.md (reported line 85)May include surrounding context.

md
注意事项:

- 必须是**子目录 + SKILL.md** 的形态(`skills/xianzong-travel-aipay/SKILL.md`),不能直接把 `SKILL.md` 放在 `skills/` 下。
- `.vscode/` 在本仓库被 gitignore,所以这份登记**只对本机生效**;新环境 clone 后需在设置里再加一改才能自动触发。
- 不要用软链指向别处(发现逻辑会做 realpath 包含性校验)。
- 改完设置后需重载窗口(Reload Window)才会重新扫描目录。

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% 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/install-and-pack.md (reported line 135)May include surrounding context.

md
| 项 | 当前值 |
|---|---|
| 资源地址 | `https://api.riyuexinzhu.com/aipay/travel_consultant` |
| 单次价格 | `0.99` 元/次 |
| 商品名 | `仙踪问道_旅行咨询` |
| 服务标识 | `API_7D36D49749854F1F` |

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 30)May include surrounding context.

md
amount_cny: "0.99"
    goods_name: "仙踪问道_旅行咨询"
    service_id: "API_7D36D49749854F1F"
  endpoint: "https://api.riyuexinzhu.com/aipay/travel_consultant"
---

# Xianzong Wendao · Travel Consultation

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_cn.md (reported line 26)May include surrounding context.

md
amount_cny: "0.99"
    goods_name: "仙踪问道_旅行咨询"
    service_id: "API_7D36D49749854F1F"
  endpoint: "https://api.riyuexinzhu.com/aipay/travel_consultant"
---

# 仙踪问道 · 旅行咨询

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/402-travel-flow.md (reported line 56)May include surrounding context.

md
alipay-bot 402-buyer-pay \
  --session-id '<当前会话标识>' \
  --file '.xianzong-travel-aipay/payment-needed.txt' \
  --resource-url 'https://api.riyuexinzhu.com/aipay/travel_consultant' \
  --method 'POST' \
  --data "$(cat '.xianzong-travel-aipay/body.json')" \
  --header 'Content-Type: application/json' \

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/402-travel-flow.md (reported line 90)May include surrounding context.

bash
alipay-bot 402-query-payment-status \
  --out-shake-no '<32 位订单号>' \
  --resource-url 'https://api.riyuexinzhu.com/aipay/travel_consultant' \
  --method 'POST' \
  --data "$(cat '.xianzong-travel-aipay/body.json')" \
  --header 'Content-Type: application/json' \

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 · scripts/travel_aipay.mjs (reported line 42)May include surrounding context.

js
import path from 'node:path';

// 资源地址与服务登记信息
const RESOURCE_URL = 'https://api.riyuexinzhu.com/aipay/travel_consultant';
const GOODS_NAME = '仙踪问道_旅行咨询';
const PRICE_CNY = '0.99';