ClawHub
Back to skill

Security audit

Phy Research Deep

Security checks for vulnerabilities and agentic risk

Overview

This is a web research workflow skill with broad activation wording, but its behavior is disclosed, purpose-aligned, and does not install code or request persistent access.

Install this if you want an agent to use web search for multi-source research. Be aware that broad phrases like "research" or "investigate" may activate it, so avoid using it for local-only or simple lookup tasks when you do not want external searches.

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 (2)

Vague Triggers

Medium
Confidence
97% confidence
Finding
The description says the skill should be used for broad natural-language requests like "research" and "investigate," which are common in normal conversation and not tightly scoped to explicit user intent for this skill. That increases the chance of accidental invocation, causing the agent to perform web searches and fetch external content when the user may have intended a simpler or local-only task.

Vague Triggers

Medium
Confidence
98% confidence
Finding
The manifest trigger list contains highly generic phrases such as "research," "investigate," and "find out about," which have high collision potential with ordinary user requests. In an agent environment, this can misroute many tasks into this skill, leading to unnecessary external search, unintended data exposure in queries, higher tool usage, and incorrect prioritization over safer local-analysis paths.

Static analysis

No suspicious patterns detected.