ClawHub
Back to skill

Security audit

guest-circuit

Security checks for vulnerabilities and agentic risk

Overview

This skill performs disclosed public podcast-appearance research and writes a visible report, with no evidence of hidden, destructive, credential-seeking, or persistent behavior.

Before installing, expect it to spend web-search calls and create a local Markdown report. Use it for guest-booking research on public information, and confirm the target person and output path when names or filenames may collide.

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
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger phrase includes broad natural-language requests like "research X before outreach," which can overlap with general research tasks and cause the skill to be invoked when the user did not specifically ask for a podcast-circuit sweep. Because the skill then performs multi-source searches and spends web-search calls, this can lead to unintended tool use, unnecessary data collection, and user-surprising behavior.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill instructs the agent to write a report to circuit/{name-slug}-{YYYY-MM-DD}.md without requiring explicit user consent or warning about file creation and possible overwrite behavior. In agent environments with filesystem access, implicit writes can create, replace, or clutter files in unexpected locations, which is a safety and integrity risk even if the content itself is harmless.

Content

No source excerpt is available for this finding.