Back to skill

Security audit

sector-returns-vs-sp500

Security checks for vulnerabilities and agentic risk

Overview

The skill provides a clearly disclosed external financial-analysis workflow and requires explicit approval before paid use.

Your question will be processed through Finch and Powerdrill, so omit confidential financial or personal information. Paid runs require your explicit approval. Review the external setup instructions before connecting Finch.

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 the user's question to an external service without a privacy warning.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The skill instructs the agent to collect the user's question and create a Finch task with that input for the Powerdrill Finance Data Agent. Although it names the services, it does not warn that the submitted question leaves the conversation or advise against including sensitive financial or personal information.

Content

No source excerpt is available for this finding.