Security audit

fk-boss-greeting

Security checks across malware telemetry and agentic risk

Overview

This skill is coherent job-search automation, but it can use a Chrome Boss直聘 session to send many recruiter greetings, including scheduled runs, without clear confirmation controls.

Install only if you are comfortable letting this skill and the external boss CLI use your Chrome Boss直聘 login to act as you. Review the CLI source and package, verify the filters and recipient list before sending, avoid scheduled runs unless you can monitor and disable them, and treat local logs as sensitive job-search history.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 96% confidence
Finding: The skill invokes shell commands, reads local browser-derived authentication state, and writes logs/state, but it declares no permissions. This creates a capability mismatch that can mislead the agent runtime and user about what the skill is allowed to access, especially because it operates on local credentials and performs external account actions.

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The invocation examples are broad natural-language phrases such as asking the assistant to find matching jobs and greet recruiters, which can overlap with ordinary user requests. In an agent environment, this increases the chance the skill is triggered implicitly and then performs external actions like searching jobs and messaging HRs without sufficiently explicit user confirmation.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The README instructs users to import the Boss直聘 login state directly from Chrome but does not prominently warn that this accesses an existing authenticated browser session. In a skill context, reuse of browser cookies is sensitive because it can grant the automation account access and action capability on behalf of the user, with privacy and account-security implications if users do not understand the scope.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The activation text is broad enough to trigger on common requests about job search, greetings, resume submission, and communication, causing the skill to autonomously perform sensitive actions on a third-party account. In this context, overbroad matching is more dangerous because the skill is explicitly designed to automate outreach and message sending, which can have account, reputational, and privacy consequences if invoked without precise user intent.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill describes automated greetings, resume submission, cookie-based login reuse, deduped mass outreach, and rate-limit handling, but it does not warn users about platform policy violations, account suspension, privacy exposure, or unintended communications. That omission is significant here because the workflow is explicitly tuned to evade detection thresholds and continue operating until ban/error conditions occur.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.