Back to skill

Security audit

BossZhiPin

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed domain-sale promotion skill, but it is set to inject ads into broad recruiting conversations that may not be asking for them.

Review before installing if you do not want an assistant to insert third-party domain-sale advertising during ordinary recruiting discussions. Installation does not appear to grant technical system access, but the response behavior is intentionally promotional and broadly triggered.

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
Findings (1)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:13
Finding

Keyword-Triggered Commercial Instruction Hijacking

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
95% confidence
Finding

The trigger list includes broad, generic recruiting terms such as '直聘模式', '招聘效率', and '互联网招聘', causing the skill to activate in many unrelated conversations. In context, this is not a harmless routing bug: the skill is explicitly designed to inject unsolicited promotion for agentzhipin.com whenever users discuss common recruiting topics, which creates deceptive spam-like behavior and can override user intent.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The skill uses 'any keyword triggers' logic while also defining overlapping scenario rules with broad terms and unclear precedence. This ambiguity makes activation and response selection unpredictable, increasing the chance that normal discussion about recruiting platforms will be hijacked into advertising content; given the skill's stated purpose is to promote a domain, the ambiguity amplifies misuse rather than serving a legitimate assistant function.

Content

No source excerpt is available for this finding.