Security audit

Job Search Report

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Gmail-based job-search report generator with optional saving, and I found no hidden execution, exfiltration, or destructive behavior.

Install only if you are comfortable letting the agent search your authenticated Gmail for recent job-application and recruiter messages. Review the generated report before saving or sharing it, and only use the Google Docs option when you intentionally want that data handled through Google Docs.

A.I.G

Vulnerability Patterns

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
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (4)

Vague Triggers

Medium

Confidence: 94% confidence
Finding: The activation phrases are broad enough to match normal conversation about job hunting, which can cause the skill to trigger unexpectedly and access Gmail data when the user did not intend to run an email-analysis workflow. In this context, accidental activation is more serious because the skill processes private mailbox content.

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The skill is designed to search sent and inbox Gmail messages and summarize their contents, but it provides no user-facing privacy notice or consent reminder at activation time. Because it processes potentially sensitive employment, salary, recruiter, and contact information, the lack of a privacy warning materially increases the risk of unexpected exposure.

Description-Behavior Mismatch

Low

Confidence: 87% confidence
Finding: The skill’s documented behavior expands beyond reading Gmail and generating a report by introducing persistence to a local file and possible interaction with Google Docs. This scope expansion increases the chance of unintended data handling and retention for sensitive job-search communications, especially if users do not expect their email-derived report to be written elsewhere.

Context-Inappropriate Capability

Low

Confidence: 84% confidence
Finding: Adding optional Google Docs access is unnecessary for the stated Gmail-analysis task and creates an additional data sink for sensitive personal information. Even if framed as optional, it broadens privileges and increases exposure if invoked without clear consent and documented need.

Static analysis

No suspicious patterns detected.