Back to skill

Security audit

wheel-strategy-options

Security checks for vulnerabilities and agentic risk

Overview

The skill sends user-provided financial details to a clearly named analysis service and explicitly limits payments and wallet access.

Your supplied financial details go to Finch and the selected agent. Share only the requested inputs and omit account identifiers and secrets. Review [Finch's privacy policy](https://www.finchtech.ai/story/privacy-policy), which describes retaining agent inputs and potentially sharing them with underlying providers. Review connection permissions separately, and approve any paid run deliberately.

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)

Financial information is sent to Finch without an explicit user-facing privacy warning.

Medium
Category
Output Handling
Confidence
93% confidence
Finding

The workflow collects available investment cash or stock holdings and cost basis, then directs the assistant to send that information to Finch without first disclosing the data transfer and obtaining informed consent. These details can reveal sensitive financial circumstances, particularly if linked to an account. Finch is openly identified as the intended service and the inputs support the requested analysis, reducing suspicion of malicious exfiltration but not resolving the privacy gap.

Content

No source excerpt is available for this finding.