ClawHub
Back to skill

Security audit

Multi Search

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed research-report generator that uses web retrieval, subagents, and local Markdown output in ways that match its stated purpose.

Before installing, expect the skill to run multiple web-research agents and create Markdown research files under an automatically selected project or research directory. Use it with material you are comfortable having summarized and researched, and specify an output path when file placement matters.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly instructs the agent to detect or create output directories and write multiple research files, but it does not clearly warn the user that local filesystem writes will occur or require explicit confirmation before doing so. In an agent environment, silent file creation can cause unintended modifications to the workspace, overwrite user-organized content, or place sensitive derived content in unexpected locations.

Natural-Language Policy Violations

Low
Confidence
89% confidence
Finding
This markdown file contains user-facing section headings in Chinese, indicating a language choice that is imposed by the skill documentation rather than offered as an option. Under the policy rule, forcing a specific language without opt-in can be a natural-language policy violation when no locale justification is provided.

Static analysis

No suspicious patterns detected.