Back to skill

Security audit

获客探索者

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill benign with medium confidence, while noting that it offers capabilities beyond its stated comment-based lead discovery purpose and that parts of the scan were incomplete.

The verdict is benign, but it is not a guarantee that every behavior was fully assessed: the scan was partial. Before installing, consider whether you want the extra account, creator-post, and trending features, and whether you are comfortable sending links or IDs to the third-party service. The scanner findings do 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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (54)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The code’s primary operation is user/account discovery, evidenced by createUserSearchTask and getUserSearchTask. Its help text explicitly distinguishes user search from video/content search. This is materially different from the declared content-search-to-comment-extraction workflow. Although account discovery could support lead generation broadly, it is not the capability described. Token handling, validation, and output logging are supporting implementation details, not separate mismatches.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

所示代码的核心调用为 detail.createDetailTask 和 detail.getDetailTask,输入为作品链接或ID,结果为作品详情。它没有调用搜索或评论相关功能,也没有对评论者数据进行分析。获取作品详情可能作为更大工作流的辅助步骤,但就所提供代码块而言,其主要行为与声明的评论区获客功能存在实质差异。API令牌校验及结果输出属于实现细节,本身不构成额外不匹配;被导入模块的内部行为未在此代码块中展示。

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

代码的主要功能是获取并输出抖音热榜,属于声明中未包含的独立能力。该入口明确拒绝所有非帮助参数,未执行关键词搜索、按内容ID获取评论,也未提取评论者信息或需求表达。令牌校验和输出封装只是支持性实现细节,不构成额外不匹配。结论仅针对所提供的代码片段。

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The supplied module calls /api/douyin/user-search/keyword and /api/douyin/user-search/info with a keyword and limit. It neither searches video content nor accepts content IDs to retrieve comments, and it does not extract demand expressions from comments. Direct user-profile discovery targets a different resource and constitutes an undeclared capability. This assessment is limited to the supplied code chunk; other parts of the skill may implement the declared workflow.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The visible command’s primary operation is creator-post retrieval, which is materially different from the declared keyword-search and comment-based lead-mining workflow. Its inputs and returned resource types concern creators and their posts, not comments or extracted leads. Token handling, validation, and output logging are supporting implementation details and are not independently treated as mismatches. This assessment is limited to the supplied chunk; imported API implementations are not shown.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The visible commands target user-search-cli.js, and their accompanying text identifies account discovery as the purpose. This materially differs from searching content and mining its commenters for leads. No content-ID lookup, comment retrieval, or commenter-field extraction is shown. The underlying CLI implementation was not supplied, so this assessment is limited to the documented invocations.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The supplied module calls /api/douyin/post/url and /api/douyin/post/info and returns post and user data. These resources and the creator-homepage input differ materially from the declared keyword-search and content-ID-based comment workflow. The chunk contains no keyword search, comment retrieval, or lead-field extraction. Task polling and response validation are supporting implementation details, not separate mismatches. This assessment is limited to the supplied code.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

该片段是CLI用法文档而非底层实现,但其明确展示的功能面向博主作品列表与账号统计,不涉及声明中的关键词搜索、按内容ID获取评论或提取评论者需求。尤其是独立获取博主账号统计的功能,不能仅视为评论区线索挖掘的实现细节,因此存在能力与主要操作对象不一致。判断仅限于所提供片段。

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

视频搜索、评论获取和用于定位或筛选内容的视频详情查询与声明用途相符。但用户搜索、博主作品采集、账号统计查询和实时热榜属于额外的独立功能,并非实现所声明评论线索挖掘流程所必需的支持性细节,因此存在能力范围不匹配。该判断基于 README 展示的命令接口,不代表已验证其底层执行行为。

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

Content search and comment retrieval match the stated purpose, and video-detail retrieval can reasonably support that workflow. However, the commands also expose user search, profile-post retrieval, and hot/trending-content discovery as standalone capabilities not disclosed in the description. These broaden the skill beyond the declared comment-based lead-mining workflow. This assessment is limited to the visible command invocations, not unprovided CLI internals.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

The supplied function retrieves a trending-search list rather than searching content by keyword or collecting and analyzing comments. Trending-topic retrieval is a distinct user-facing capability, not merely request handling or another supporting implementation detail. The assessment is limited to the supplied code chunk.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding

该片段的明确用途是获取视频详情,未展示评论区访问或线索提取,也未说明详情查询如何作为声明流程的辅助步骤。因此,就所提供的片段而言,其主要操作与声明用途不一致。由于未提供 detail-cli.js 的实现,无法判断该脚本是否另有相关功能或其他行为。

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 76)May include surrounding context.

md
- '抖音上"AI 教程"最火的视频 → node src/douyin/search-cli.js --keyword "AI 教程" --sort 1'

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 77)May include surrounding context.

md
- '最近一周关于"AI 模型"的最新内容,要 20 条 → node src/douyin/search-cli.js --keyword "AI 模型" --sort 2 --time 7 --limit 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 84)May include surrounding context.

md
- '按关键词在评论区找高意向客户 → node src/douyin/search-cli.js --keyword "装修" --sort 1 --limit 20,再用 aweme_id 跑 comment-cli,结束后读 references/customer-acquisition.md'

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 88)May include surrounding context.

md
- '帮我找 1 分钟以内的减肥短视频,先来 30 条 → node src/douyin/search-cli.js --keyword "减肥" --duration 1 --limit 30'

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 89)May include surrounding context.

md
- '半年内最火的 AI 教程图文 → node src/douyin/search-cli.js --keyword "AI 教程" --sort 1 --time 180 --content 2'

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 128)May include surrounding context.

md
| 搜视频/图文、爆款、选题 | `node src/douyin/search-cli.js` | `--keyword` |

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 144)May include surrounding context.

md
node src/douyin/search-cli.js --keyword "AI 教程" --sort 1 --limit 10

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 78)May include surrounding context.

md
- '找做 AI 的对标账号 / 搜达人 → node src/douyin/user-search-cli.js --keyword "AI" --limit 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 110)May include surrounding context.

md
| 运营 / 投放 | 按关键词找对标账号、达人 | `user-search-cli.js` |

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 129)May include surrounding context.

md
| 搜用户/达人/账号、按词找对标 | `node src/douyin/user-search-cli.js` | `--keyword` |

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 145)May include surrounding context.

md
node src/douyin/user-search-cli.js --keyword "AI" --limit 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 79)May include surrounding context.

md
- "今天抖音有什么热点 / 热搜榜 → node src/douyin/hot-cli.js"

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
| 所有人 | 今天抖音在火什么 | `hot-cli.js` |