ClawHub
Back to skill

Security audit

Kite Agent Smart Wallet Permissionless Protocol

Security checks for vulnerabilities and agentic risk

Overview

This skill is coherent documentation for AI smart wallets, but its examples can authorize wallet actions and session keys without enough safety scoping for real funds.

Install only if you understand Kite smart-wallet operations. Treat the snippets as potentially asset-affecting: verify chain ID and contract addresses, prefer testnet or low-value accounts first, set narrow session-key limits and allowed functions, and confirm how to revoke keys before using mainnet.

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
92% confidence
Finding
The README includes example code for creating a wallet and adding a session key, both of which are on-chain permission-granting actions, but it provides no warning about transaction signing, gas costs, delegated authority, or the security implications of session keys. In the context of an AI-agent smart wallet protocol, omission of these warnings can lead users to grant broad authority or deploy wallets without understanding the resulting permissions and financial risk.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The usage examples show wallet creation, session-key addition, and transaction execution flows without any explicit warning that these operations can move assets, create privileged credentials, or be pointed at mainnet endpoints listed earlier in the document. In an agent-skill context, operators may copy these snippets directly or allow an LLM-driven agent to operationalize them, increasing the chance of unintended fund movement or unsafe permission delegation.

Natural-Language Policy Violations

Low
Confidence
93% confidence
Finding
The architecture diagram contains Chinese text while the rest of the README is in English, which imposes a mixed-language presentation without opt-in or explanation. This can violate language/locale consistency expectations for organizational documentation.

Static analysis

No suspicious patterns detected.