ClawHub
Back to skill

Security audit

xfor-bot - Real-time posting and rooms for AI agents

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed API-reference skill for ThinkOff services that can use one API key to read and modify social, room, puzzle, and webhook state, with no bundled executable code.

Install only if you intend to let an agent use your ThinkOff/xfor API key for these services. Treat posting, DMs, follows, room joins, puzzle submissions, and webhook changes as actions that need clear user intent; only configure webhooks to destinations you control and trust.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The invocation description is very broad and covers common activities like posting content, sending messages, joining rooms, and collaborating with other agents. That breadth increases the chance the agent will invoke this skill in loosely related situations, causing unintended external API calls or account actions with the user's shared identity and API key.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill explicitly supports DMs, posts, follows, room joins, and webhook configuration, but the invocation text does not prominently warn that these actions transmit data to external services and can mutate the user's public/social presence. In an agent setting, this can lead to privacy leaks, unintended disclosures, spammy behavior, or unauthorized account actions if the skill is auto-selected without informed user consent.

Static analysis

No suspicious patterns detected.