Back to skill

Security audit

XH Agents x402

Security checks for vulnerabilities and agentic risk

Overview

This skill is a clearly disclosed paid x402 client that can spend USDC from a wallet key you provide, with no evidence of hidden persistence or unrelated data access.

Install this only where it is acceptable for an agent-operated tool to read a dedicated funded Base wallet key and spend USDC. Use a low-balance wallet, keep the default or a tight --max value, use --pay-to for known destinations, and do not submit private documents, internal URLs, or regulated data to the paid routes unless external processing by xhagents.xyz is approved.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The document extraction and web retrieval capabilities imply that user-supplied files and fetched page contents may be sent to an external paid API, but the route list gives no warning about that data flow. This can lead users or agents to submit sensitive documents or retrieve confidential internal URLs without realizing the content may leave their environment and be retained or logged by a third party.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

These routes advertise workflows involving connecting a headless VPS to a user's Google Drive and automating YouTube upload activity, which inherently involve account access, tokens, and potentially personal or proprietary data. Presenting them without any warning about credential handling, consent, token scope, storage, or privacy increases the risk that an agent or user will invoke them without understanding that sensitive account data may be transmitted to or processed by external services.

Content

No source excerpt is available for this finding.

Dynamic attribute access via getattr()

Low
Category
Dangerous Code Execution
Confidence
50% confidence
Finding

Dynamic getattr() with a non-literal attribute name can access arbitrary object attributes, potentially bypassing access controls.

Content

Scanner excerpt · scripts/xh_pay.py (reported line 133)May include surrounding context.

python
if obj is None:
        return {}
    for attr in ("model_dump", "dict"):
        method = getattr(obj, attr, None)
        if callable(method):
            try:
                return method()