职业规划 + 简历生成 + 求职优化

• Outcome: Pass
• Latest audit: Jul 31, 2026 · 7:23 AM UTC
• Version: 0.1.0

Overview

This career and resume skill is coherent and purpose-aligned, but users should be mindful that it stores sensitive career data locally and can share it to Feishu when asked.

Install only if you are comfortable storing career, assessment, and resume details in local ./career and ./resume files. Before using Feishu sharing, confirm exactly what will be uploaded and who will receive it, and keep a backup before letting the skill rewrite resume.yaml.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (9)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 90% confidence
• Finding: The skill explicitly documents reading and writing local files such as `./career/profile.yaml`, `./resume/resume.yaml`, and output artifacts, yet no permissions or user-consent model are declared. Undeclared file access increases the risk of silent data handling, especially because resumes and career plans contain sensitive personal information and the skill can create or overwrite files.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 78% confidence
• Finding: The documented behavior does not fully match the described purpose, including additional assessment/export/validation capabilities and overstating bullet rewriting as if it performs full rewriting when it may only diagnose. Capability mismatches can mislead users and reviewers about what data is processed and what actions occur, weakening informed consent and security review accuracy.

Vague Triggers

Medium

• Confidence: 83% confidence
• Finding: The routing triggers are broad phrases like writing resumes, career planning, or optimization requests, which can match ordinary conversation and cause unintended module activation. In this skill, unintended routing is more dangerous because downstream modules may read or write personal files or export artifacts without the user clearly requesting those actions.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The shared workflow states that the skill will create and populate career and resume files and generate HTML/PDF outputs, but the user-facing description does not warn about these write operations. This reduces transparency around storage and artifact generation for highly sensitive employment data, making accidental persistence or disclosure more likely.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The module instructs publishing career reports to Feishu and embedding HTML visualizations without warning that sensitive personal data, assessment results, and career plans will be sent to an external service. This creates a real privacy and data-governance risk because users may unknowingly disclose educational history, psychometric outputs, and career intentions to third-party platforms.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The file explicitly instructs writing each round of collected profile data directly into `profile.yaml` for session persistence, while the data categories include personal and potentially sensitive information such as name, age, city, education background, family constraints, health status, GPA, and employment history. Persisting this data without any notice about consent, minimization, retention, access controls, or deletion creates a real privacy and security risk if the data is stored insecurely, over-collected, or retained longer than necessary.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The module explicitly instructs the agent to generate and later modify `resume.yaml` in the user's working directory, but it does not require clear disclosure or confirmation before writing files. In an agent setting, silent file creation/modification can surprise users, overwrite existing content, or cause unintended persistence of sensitive resume data on disk.

Missing User Warnings

High

• Confidence: 96% confidence
• Finding: The Feishu-sharing workflow directs the agent to convert resume data and send it through `lark-doc`/`lark-im`, but it does not require a privacy warning or explicit consent acknowledging that personally identifiable information will be transmitted to an external service. Because resumes contain highly sensitive personal and professional data, external sharing without informed consent creates a substantial privacy and data-exposure risk.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The module explicitly says the agent will directly update resume.yaml after bullet rewrites and ATS fixes, but it does not require explicit user confirmation, preview, backup, or change summary before modifying user data. This creates an unsafe write behavior where an agent could overwrite or silently alter a user's resume content, potentially causing data loss, unwanted edits, or integrity issues in a sensitive job-application document.

VirusTotal

63/63 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.