Security audit

第一阶段：批量筛选简历

Security checks for vulnerabilities and agentic risk

Overview

This resume-screening skill is mostly purpose-aligned, but it should be reviewed because it handles sensitive hiring data and instructs agents to delete prior report files without a confirmation step.

Install only if you are authorized to process the candidate resumes and can store the generated spreadsheets according to your hiring data policy. Use a dedicated output folder, review scoring criteria before relying on recommendations, and do not allow the delete-old-reports step unless the exact files are shown and confirmed or backed up first.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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)

Context-Inappropriate Capability

Medium

Confidence: 97% confidence
Finding: The skill includes explicit shell commands to delete existing report files before regeneration, which introduces destructive filesystem behavior unrelated to the core business purpose of screening resumes. In an agent setting, broad path substitution or user-controlled target directories could cause unintended or irreversible data loss beyond the generated artifacts.

Missing User Warnings

High

Confidence: 98% confidence
Finding: The skill instructs deletion of pre-existing output files without any explicit warning, confirmation, or safety boundary. In an automated agent workflow, this can lead to silent destruction of user data, especially if the target directory is misconfigured, reused for other files, or influenced by external input.

Static analysis

No suspicious patterns detected.