ClawHub
Back to skill

Security audit

code-search

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small GitHub code-search guide with no executable payload, but its broad triggers may activate it more often than users expect.

Install only if you are comfortable with broad trigger words potentially routing ordinary code or search requests into GitHub-oriented web research. The artifact does not show hidden execution, persistence, credential use, or local file access, but more specific triggers would be preferable.

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 (3)

Shadow Command Trigger

Medium
Category
Trigger Abuse
Confidence
95% confidence
Finding
The trigger "search" is likely to conflict with built-in search functionality because it is a highly generic command word. In practice, this can cause command shadowing, ambiguous routing, and unintended invocation of this skill instead of the platform’s native search behavior.

Vague Triggers

Medium
Confidence
98% confidence
Finding
The trigger list includes broad single-word activators like "github", "code", and "search", which can cause the skill to activate in many unrelated contexts. Because this skill directs external research behavior, unintended activation can misroute user requests, shadow more appropriate tools, or cause unnecessary external lookups.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
97% confidence
Finding
Using "code" as a trigger is overly broad because it appears in a large fraction of developer conversations, even when the user is not asking to search GitHub. This increases accidental invocation risk and may divert the agent from safer or more relevant local analysis tools.

Static analysis

No suspicious patterns detected.