Back to skill

Security audit

Defi Intelligence Skill

Security checks for vulnerabilities and agentic risk

Overview

This skill is a clearly disclosed DeFi gateway wrapper with real-money and privacy risks, but it requires human approval for irreversible on-chain actions.

Install only if you trust gateway.spraay.app and are comfortable paying USDC per request and sending wallet addresses, portfolio data, and contract call details to that service. Use a limited-purpose wallet and manually verify every write transaction before approving it.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 164)May include surrounding context.

md
### ⚠️ Execution (WRITE — requires human approval)

> **These endpoints submit irreversible blockchain transactions.** The agent MUST request explicit human confirmation before calling either endpoint. Show the user the full parameters (token, amount, recipient, chain, slippage) and wait for approval. Never auto-execute.

**Execute Swap** — $0.015 ⚠️ WRITE
Execute a token swap on Base via the MangoSwap router. **Use swap/quote first** to preview the trade. Tokens will leave the wallet.