Back to skill

Security audit

airdrop-eligibility-tracker

Security checks for vulnerabilities and agentic risk

Overview

This is a read-only wallet activity checker that discloses its public RPC network use, with some scoring-description caveats users should understand.

Install only if you are comfortable sending public wallet addresses to publicnode.com RPC endpoints. Treat the score as a rough activity indicator, not a real eligibility prediction, because the implementation does not verify project-specific criteria, account age, recency, bridge use, or distinct contract interactions.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (5)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding

The skill performs outbound network access to public RPC endpoints but does not declare a corresponding tool scope or permissions boundary in the skill manifest. That creates a transparency and control gap: an agent or platform may invoke network-capable behavior without clear operator awareness or policy gating, increasing the chance of unintended external requests and data disclosure of user-supplied wallet addresses.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The invocation guidance includes broad user phrasing like checking whether a wallet 'looks active enough' or asking how airdrop farming is going, which could match normal conversation and cause the skill to trigger more often than intended. In a network-enabled skill, overbroad triggering can lead to unnecessary external RPC calls and inadvertent sharing of wallet addresses with third-party infrastructure.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The module documentation says the tool scores using 'wallet age via nonce' and 'account age' and references 'distinct contracts touched'. In the implementation, the score is derived only from total transaction count, active chain count, and whether any chain balance exceeds 0.01; no age, recency, or distinct-contract metric is computed, so the documentation materially overstates what the score reflects.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The script sends queried wallet addresses to third-party public RPC providers, which exposes user interest and wallet lookups to external services without an explicit privacy notice. While blockchain addresses are public, correlating them with a user's IP, timing, and usage patterns can leak metadata and reduce privacy, especially in a tool aimed at comparing multiple wallets.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The comments state the score is modeled on criteria including cross-chain or bridge activity and mention diminishing returns with bot-related rationale. The actual code only applies simple caps to total transaction count and active-chain count, with no recency metric, no bridge detection, and no bot/sybil-specific adjustment, creating a misleading explanation of what the scoring logic does.

Content

No source excerpt is available for this finding.