T09 · Insecure Skill Coding Practices
Error
- Location
SKILL.md:59- Finding
Chromium Sandbox Disabled for Authenticated Browser Automation
- Content
View full analysis
- Remediation
View remediation
Security audit
Linkedin Easy Apply Automation
Security checks for vulnerabilities and agentic risk
Overview
The skill openly automates LinkedIn Easy Apply submissions, but it gives an agent live application authority and recommends an insecure Chromium launch configuration for an authenticated browser profile.
Review this carefully before installing. Use a dedicated browser profile and preferably a separate OS account or container, keep DRY_RUN enabled until you inspect targets and answers, require manual approval before each live submission, and remove the Chromium sandbox-disabling flags unless your environment has a well-understood reason.
Vulnerability Patterns
- Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
- Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
- Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
- Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
- Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)
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 (2)
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 91% confidence
- Finding
The skill's 'when_to_use' language is broad enough to trigger in generic job-application scenarios, even though it performs high-impact real-world actions on third-party platforms. That increases the chance an agent invokes it without sufficiently explicit user consent or without the user appreciating that actual applications may be submitted.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The skill describes automation and submission behavior but does not prominently warn that it can submit real job applications on the user's behalf to external employers. In this context, omission of that warning is dangerous because users or calling agents may treat the skill like a search assistant rather than an action-taking system, leading to unintended submissions, reputational harm, and account/platform consequences.
- Content
Static analysis
No suspicious patterns detected.