Auto Job Applying Agent

• Outcome: Review
• Latest audit: May 29, 2026 at 9:34 PM UTC
• Version:
  2.0.1

Overview

This skill mostly does what it claims, but it needs review because it can submit real job applications with personal data and contains a misleading browser-automation disclosure.

Review before installing. Keep AUTO_APPLY_MODE=false, approve only specific jobs, use a dedicated revocable Resumex API key, avoid Telegram delivery unless needed, and install dependencies in a virtual environment. Be aware that submitted applications and tracker logs may be difficult or impossible to undo.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (7)

Intent-Code Divergence

High

• Confidence: 97% confidence
• Finding: The privacy notice says no outbound calls are made other than Resumex and optional Telegram, yet the documented auto-apply flow explicitly submits resume data to third-party job portals through browser automation. This is dangerous because users may consent under a materially false understanding of where their personal data will be transmitted.

Intent-Code Divergence

Medium

• Confidence: 95% confidence
• Finding: The docstring explicitly claims the browser is transparent and detectable, but the implementation later launches Chromium with `--disable-blink-features=AutomationControlled`, which suppresses a common automation signal. In a job-application automation skill handling sensitive personal data, this mismatch is security-relevant because it misrepresents actual bot-evasion behavior to users and reviewers, undermining informed consent and trust assumptions.

Intent-Code Divergence

Medium

• Confidence: 98% confidence
• Finding: The inline comment states there is no stealth or bot-detection evasion, yet the launch arguments include `--disable-blink-features=AutomationControlled`, an anti-detection measure. This is dangerous because the skill automates submissions to third-party job portals while presenting itself as honest and non-evasive, creating deceptive behavior that may violate platform expectations and expose users to account or application risks.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The script transmits resume contents, which commonly include sensitive personal data such as name, email, phone, location, employment history, and education, to Telegram once bot credentials are present. Although this appears to be the feature's intended behavior, there is no explicit consent checkpoint or warning at send time, which increases the risk of accidental disclosure to a wrong chat, compromised bot, or unintended third-party platform.

Unpinned Dependencies

Low

• Category: Supply Chain

Content

#
# Approx. disk usage after install: Playwright package ~10MB, Chromium binary ~300MB.

playwright>=1.42.0
requests>=2.31.0

• Confidence: 95% confidence
• Finding: playwright>=1.42.0

Unpinned Dependencies

Low

• Category: Supply Chain

Content

# Approx. disk usage after install: Playwright package ~10MB, Chromium binary ~300MB.

playwright>=1.42.0
requests>=2.31.0

• Confidence: 98% confidence
• Finding: requests>=2.31.0

Known Vulnerable Dependency: requests — 10 advisory(ies): CVE-2014-1830 (Exposure of Sensitive Information to an Unauthorized Actor in Requests); CVE-2024-47081 (Requests vulnerable to .netrc credentials leak via malicious URLs); CVE-2024-35195 (Requests `Session` object does not verify requests after making first request wi) +7 more

High

• Category: Supply Chain
• Confidence: 91% confidence
• Finding: requests

VirusTotal

67/67 vendors flagged this skill as clean.