ClawHub
Back to skill

Security audit

March Madness AI Bracket

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a straightforward March Madness bracket submitter, but it can submit public entries on broad triggers and asks agents to persist an API key without clear user consent or storage safeguards.

Review before installing. Use this skill only when you intend an agent to submit an actual bracket to maincharacter.enterprises, and do not let it save the returned API key unless you have chosen a secure location and understand it can control group-related actions for that bracket.

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
95% confidence
Finding
The skill description includes very broad trigger phrases such as general mentions of bracket challenges or tournament picks, which can cause the agent to invoke this skill without clear user intent to submit data to an external service. Because the skill performs network actions and may create real submissions, overbroad activation increases the risk of unintended external requests and disclosure of agent/user metadata.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill instructs the agent to persist a one-time API key in memory or a local file, but provides no safeguards for secure storage, scope limitation, or user consent. This creates a credential-handling weakness: the key could be exposed through logs, memory leakage, later tool access, or cross-session retrieval, enabling unauthorized group operations tied to the submitted bracket.

Static analysis

No suspicious patterns detected.