Vietnam Fullstack Jobs

• Outcome: Pass
• Latest audit: May 28, 2026 · 6:18 AM UTC
• Version: 1.0.1

Overview

This is a straightforward job-search helper that uses Firecrawl to search public Vietnam tech job sites, with the main caution being automatic external API usage.

Install this only if you want your agent to run Firecrawl-backed searches against public job sites when job-search prompts match this skill. Use a dedicated, revocable Firecrawl API key, watch quota or billing, and review the required firecrawl-search skill separately.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Vague Triggers

Medium

• Confidence: 85% confidence
• Finding: The README advertises very short, generic trigger phrases such as "job mới" and "fullstack jobs," which can cause the skill to activate on broad employment-related requests without clear boundaries. In an agent setting, ambiguous activation can lead to unintended invocation, wasted tool usage, and unexpected access to external search capabilities, though the direct security impact here is limited because the described behavior is job search rather than privileged action.

Vague Triggers

Low

• Confidence: 79% confidence
• Finding: The README uses non-exhaustive wording like "Trigger for requests like" without defining precise scope limits, so the agent may overgeneralize and activate the skill for adjacent requests that were not intended. Because this skill performs external job searches via Firecrawl, ambiguous activation increases the chance of unnecessary network actions and inconsistent behavior, but the domain context makes the impact relatively low.

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The trigger phrases include very broad terms like 'job mới' and 'fullstack jobs', which can match ordinary conversation and cause the skill to activate when the user did not clearly request this specific workflow. Because the skill then runs multiple searches automatically, accidental invocation can lead to unintended external requests and irrelevant task takeover.

Natural-Language Policy Violations

Medium

• Confidence: 91% confidence
• Finding: The skill forces responses in Vietnamese regardless of the user's language preference, which can override user intent and reduce transparency about what the agent is doing. This is primarily a policy and usability issue, but it can also make results less accessible or misleading for users expecting another language.

Autonomous Decision Making

Medium

• Category: Excessive Agency

Content

## Procedure

Run all search queries automatically, without asking for confirmation. Deduplicate results across all queries and sources.

**Step 1 — Search ITviec**

• Confidence: 90% confidence
• Finding: without asking

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.