ClawHub
Back to skill

Security audit

Kobe

Security checks for vulnerabilities and agentic risk

Overview

This is a read-only Kobe Bryant knowledge skill with some broad activation wording but no code, persistence, data access, or hidden behavior.

Installers should expect this skill to answer Kobe Bryant-related prompts and may want narrower activation rules if accidental triggering on #8, #24, or broad basketball references would be annoying. No security-sensitive access or persistence was found.

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
  • 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 (5)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The activation criteria are broad enough that ordinary basketball discussions mentioning 'Kobe,' '#8,' '#24,' or loosely related concepts could trigger the skill unintentionally. In a prompt-based agent system, this can cause inappropriate routing, override more relevant skills, or inject Kobe-specific responses into unrelated conversations, reducing reliability and potentially exposing users to unintended behavior.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The activation rule is overly broad because it triggers not only on explicit mentions of Kobe Bryant but also on generic references like '#24', '#8', or implied mentions. In a skill-routing system, this can cause unintended invocation on unrelated user requests, leading to incorrect responses, context hijacking, or reduced reliability of other skills.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

Using vague trigger language like 'implies' and 'related topics' creates ambiguous scope for automatic activation. That ambiguity can be exploited by ordinary conversation patterns or overlapping topics, causing this skill to activate when the user did not intend a Kobe-related response.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The usage section reinforces automatic activation from 'related keywords' without clarifying what qualifies or what should be excluded. This increases the chance of over-triggering during normal conversation, which can misroute user intent and degrade trust in the assistant's behavior.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

The activation list includes the Chinese term '小飞侠' as a trigger, but the README does not explain language handling or whether multilingual behavior is user-driven. This can create an implicit locale/language policy issue if the skill responds based on a non-default language cue without documenting user choice.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.