Back to skill

Security audit

query-fan-out-coverage

Security checks for vulnerabilities and agentic risk

Overview

The verdict is benign because the scans report no clear malicious behavior, though they flag a privacy disclosure gap and the review was incomplete.

Before installing, consider whether you are comfortable sharing business prompts and targeting details with external services. The scan was partial because some local references could not be resolved, so the evidence does not establish that every referenced item was reviewed.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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)

The workflow sends user-provided business prompts to external services without a privacy warning.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The DataForSEO workflow instructs the agent to transmit the user's money prompt, location, and language to a third-party API; L050 also directs submitting prompts to several external AI services. The skill does not warn that these submissions disclose prompt contents to those providers or advise excluding confidential business information; the paid-call approval requirement addresses cost rather than privacy.

Content

No source excerpt is available for this finding.