Security audit

Resume Assistant

Security checks across malware telemetry and agentic risk

Overview

This is a prompt-only resume helper whose sensitive data handling is expected for its purpose, with privacy and trigger-scope cautions users should understand.

Install only if you are comfortable sharing resume and job-description text with the agent or API provider running it. Redact unnecessary personal details, avoid confidential employer or third-party information, review any generated achievements for accuracy, and run export/conversion commands only on files you trust.

SkillSpector

By NVIDIA

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 (12)

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The integration section encourages sending full resume and job-description content through APIs and tool wrappers but provides no privacy or data-handling warning. Because resumes commonly contain PII and job descriptions may contain confidential recruiting information, users and integrators may unknowingly transmit sensitive content to third-party services or logs.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The README encourages 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 cause the skill to activate on ordinary conversation containing highly sensitive personal data. In a resume-processing skill, overbroad invocation increases the chance that users disclose or the agent routes personally identifiable information into processing/export flows without clear confirmation or consent.

Missing User Warnings

Medium

Confidence: 86% confidence
Finding: The README advertises multi-format export of resumes but does not warn that resumes typically contain sensitive personal and employment data that will be ingested, transformed, and possibly stored or transmitted by downstream systems. This omission can mislead integrators and users into treating export as harmless formatting rather than sensitive-data processing.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The skill encourages activation from very broad everyday phrases like 'Here's my resume, can you help?' and 'What's wrong with my resume?', which increases the chance of over-triggering in normal conversation and routing sensitive resume content into this skill without clear user intent. In agent environments with automatic intent detection, this can cause unintended processing or disclosure of personal employment data.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The documentation shows REST API usage that transmits resume and job-description text to an HTTP endpoint, but it does not warn users that highly sensitive personal data may leave the local environment. Because resumes often contain names, contact details, work history, and sometimes addresses or other identifiers, omission of this disclosure creates a meaningful privacy and compliance risk.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The skill encourages very broad natural-language triggers such as '帮我看看' or '这是我的简历，帮我看看', which can overlap with normal conversation and cause the agent to invoke resume-processing workflows without clear user confirmation. In a skill that processes sensitive personal data like resumes and job descriptions, ambiguous invocation increases the chance of unintended data handling or routing.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The routing logic maps broad keywords like '润色/优化/修改', '求职/申请/匹配', and '导出/下载/转换' directly to commands without defining exclusions, precedence, or confirmation steps. This can cause misrouting or unintended execution of a workflow, especially when the input contains sensitive resume data or mixed intents.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The documentation invites users to paste resumes and job descriptions, which often contain personally identifiable and sensitive employment data, but it does not provide a clear upfront warning to minimize or redact sensitive fields before submission. Although a later FAQ says content is not stored, that does not replace explicit privacy guidance at the point of collection or address transmission to external services.

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The usage guide promotes very broad natural-language triggers such as 'Here's my resume, can you help?' and 'What's wrong with my resume?'. In agent environments where skills are auto-routed from free-form text, these generic phrases can cause unintentional invocation, leading the skill to process sensitive resume/CV data when the user may not have explicitly intended to call this skill.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The export trigger list contains very broad phrases such as 'convert to', 'export as', and 'save as' that are not scoped to resumes. In a chat environment, these can cause the resume skill to activate for unrelated file-conversion requests, leading to incorrect tool routing, unintended processing of non-resume content, and possible exposure of sensitive user data to the wrong skill context.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The customize trigger list includes generic phrases like 'create a resume for', 'write a resume for', and especially 'target this job' or 'match this job description', which may overlap with ordinary career or document-help requests. This increases the risk of accidental invocation on ambiguous prompts, causing misclassification of user intent and potentially processing unrelated user-provided text as resume content.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: Several triggers are generic phrases such as 'convert to', 'save as', or 'create a resume for' that may match ordinary conversation without enough resume-specific context. This can cause unintended invocation of the skill on unrelated user content, potentially exposing sensitive text to resume-processing prompts or causing the agent to take the wrong action.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.