Security audit

Resume Assistant

Security checks for vulnerabilities and agentic risk

Overview

This is a prompt-only resume assistant; its broad routing and privacy guidance should be tightened, but the artifacts match the stated purpose and show no hidden execution, persistence, or exfiltration.

Before installing, use explicit commands for export or job-tailoring actions and avoid pasting unnecessary personal details. If you expose this skill through an API or hosted agent, add a clear privacy notice and obtain user consent before transmitting resumes or job descriptions outside the local chat.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (11)

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The README advertises very broad natural-language triggers such as 'Here's my resume, can you help?' and 'What's wrong with my resume?', and states that the assistant will 'automatically route' based on inferred intent. In agent environments where skills are selected from free-form user text, this increases the chance of unintended invocation, prompt-routing ambiguity, or accidental handling of sensitive resume/job-description content without the user explicitly choosing the skill or operation.

Vague Triggers

Medium

Confidence: 86% confidence
Finding: 这些自然语言触发示例非常宽泛，接近日常对话表达，容易在用户只是寻求一般性建议或展示简历内容时被误判为技能调用。对一个会自动执行润色、评分、导出等工作流的技能来说，模糊触发边界会导致非预期处理、错误路由，甚至把敏感简历内容送入不必要的后续流程。

Vague Triggers

Medium

Confidence: 90% confidence
Finding: 文档宣称“直接描述你的需求”即可自动路由，但没有定义触发边界、优先级或排除条件，这会放大意图识别错误的风险。由于该技能面向自由文本输入，缺乏明确边界可能导致普通聊天内容被当作简历处理请求，造成误操作、数据暴露范围扩大或错误输出。

Vague Triggers

Medium

Confidence: 94% confidence
Finding: The skill advertises very broad natural-language triggers such as 'Here's my resume, can you help?' and 'What's wrong with my resume?', which overlap with ordinary conversation and can cause unintended routing into the skill. In an agent that auto-loads skills based on free-form intent detection, this increases the chance that sensitive resume or job-description content is processed by this skill without an explicit user action.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The intent-routing logic relies on ambiguous keyword matches like 'job', 'apply', 'match', 'score', or 'help', without any boundary checks, contextual disambiguation, or confirmation step. This can lead to accidental invocation, misrouting, and exposure of sensitive user-provided content to resume-processing prompts or downstream integrations when the user did not intend to use this skill.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The REST API integration section shows resume and job-description content being sent to HTTP endpoints, but the skill documentation does not warn that highly sensitive personal and employment data may leave the local agent and be transmitted to external services. This creates a privacy and data-handling risk because deployers or users may unknowingly expose PII, career history, contact details, and target-job information to third-party systems.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The skill advertises very broad natural-language triggers such as '这是我的简历，帮我看看' and '我的简历有什么问题？', which can overlap with ordinary conversation and cause unintended invocation of this skill. In an agent that auto-routes based on intent, this can expose sensitive resume data to the wrong workflow or trigger processing/export actions without sufficiently explicit user intent.

Vague Triggers

Medium

Confidence: 86% confidence
Finding: The usage guide encourages broad natural-language triggers such as 'Here's my resume, can you help?' and 'Create a resume...' that can cause the skill to activate or process sensitive resume/job-description content without explicit, scoped user intent. In a skill that can score, customize, and export documents, ambiguous invocation increases the chance of unintended data handling, including processing or file-generation actions the user did not clearly request.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The guide invites users to paste resumes and job descriptions but does not warn that this content may contain highly sensitive personal and employment information or that export operations may create files in additional formats. Users may unknowingly expose PII or trigger persistence/sharing behaviors without understanding the privacy implications.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The skill defines very broad natural-language triggers such as 'create a resume for', 'convert to', and 'save as', which can match ordinary user conversation rather than a deliberate invocation of this skill. That can cause unintended routing of sensitive resume or job-description content into this skill, creating authorization, privacy, and prompt-routing risks even though the manifest itself does not contain direct code execution behavior.

Vague Triggers

Medium

Confidence: 87% confidence
Finding: The natural-language triggers include generic phrases such as 'convert to', 'save as', 'create a resume for', and 'target this job' that may be matched outside a clearly resume-scoped interaction. In an agent environment, this can cause unintended invocation of the skill on unrelated user content, potentially exposing sensitive text to resume-processing prompts or producing confusing actions the user did not intend.

Static analysis

No suspicious patterns detected.