ClawHub
Back to skill

Security audit

Kite Agent Wallet

Security checks for vulnerabilities and agentic risk

Overview

This documentation-only skill is coherent for a Kite AI smart-wallet workflow, but users should treat its wallet and session-key examples as real authorization steps.

Install only if you intend to work with Kite AI smart wallets. Use testnet or throwaway accounts first, verify all contract addresses and network settings, understand that session keys delegate authority with spending limits, and revoke keys you no longer need.

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
  • 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 (3)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The usage examples instruct users to create wallets, add session keys, and execute transactions on a blockchain network without any explicit warning that these actions may move assets, consume gas, create irreversible state changes, or grant delegated spending authority. In the context of an agent wallet protocol, this omission is risky because operators may copy and run the examples assuming they are harmless setup steps, when they can authorize real on-chain actions and persistent permissions.

Natural-Language Policy Violations

Low
Confidence
94% confidence
Finding
SQP-3 applies to natural-language policy issues in all file types, including markdown. The architecture diagram includes Chinese text (e.g. '授权', '使用', '可添加多个session key，每个有限额') in an otherwise English README, which imposes a language choice without user opt-in or documented justification.

Missing User Warnings

Low
Confidence
87% confidence
Finding
This is a markdown file, so SQP-2 applies to missing user-facing warnings in documentation. The Quick Start demonstrates creating a wallet and adding a session key with spending limits, which can affect user assets or authorization state, but the section provides no caution about using testnet vs. mainnet, verifying addresses, or the security implications of granting session-key permissions.

Static analysis

No suspicious patterns detected.