ClawHub
Back to skill

Security audit

Agent Madness Bracket Skill

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly transparent about entering a paid bracket contest, but it asks an agent to sign live USDC payments from a wallet based on remote payment instructions without strong local checks.

Install only if you are comfortable with an agent submitting data to agentmadness.fun and making a live Base USDC payment. Use a dedicated wallet funded only with the exact entry amount plus minimal gas, verify the contest is still active after the stated March 19, 2026 deadline, and do not run the payment example unless you can independently confirm the amount, token, chain, and recipient before signing.

A.I.G

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (2)

T09 · Insecure Skill Coding Practices

Error
Location
SKILL.md:210
Finding

Remote Server Payment Instructions Are Not Explicitly Validated Before Signing

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
SKILL.md:49
Finding

Security-Critical npm Dependencies Are Installed Without Version Pinning

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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)

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 183)May include surrounding context.

Step 4: Validate Picks (Free — Always Do This Before Paying)

javascript
const validation = await fetch("https://agentmadness.fun/api/validate-picks", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({

External Transmission

Medium
Category
Data Exfiltration
Confidence
83% confidence
Finding

This call initiates an x402 payment flow that causes the local wallet to sign and authorize an on-chain USDC transfer based on payment instructions from a remote server. Even though the private key is not exfiltrated, the skill drives real asset movement to an external service, so a compromised, spoofed, or malicious endpoint could induce unwanted payment or payment to unexpected terms if the client or user does not independently verify recipient, amount, chain, and token.

Content

Scanner excerpt · SKILL.md (reported line 219)May include surrounding context.

md
});

// Submit — x402Fetch handles the 402 → pay → retry automatically
const result = await x402Fetch("https://agentmadness.fun/api/submit-bracket", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({

Static analysis

No suspicious patterns detected.