Back to skill

Security audit

Deepblue Defi Api

Security checks for vulnerabilities and agentic risk

Overview

This read-only DeFi API skill is mostly purpose-aligned, but its privacy claims conflict with its IP-based limits and possible wallet-to-IP tier handling.

Review the API operator's privacy policy or source before using wallet-scan or holder-tier features. Avoid submitting wallet addresses you do not want linked to your IP address until the service clearly documents rate-limit storage, logs, retention, and whether wallet addresses can be associated with network metadata.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (2)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The documentation makes a strong privacy claim that IP addresses are not stored or logged, yet the skill also states rate limits are enforced per IP. Even if only transiently processed, per-IP enforcement necessarily involves handling IP-linked state, so the current wording is misleading and can cause users to disclose wallet activity under false privacy assumptions.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill says requests are anonymous and involve no wallet identification, but then describes associating a requesting IP with an address to enforce holder tiers. That creates an identity/linkage contradiction: users may believe requests are unlinkable when the service may correlate wallet addresses with network identifiers, exposing behavioral and financial privacy risk.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.