ClawHub
Back to skill

Security audit

Sector Returns vs S&P 500 (Powerdrill via Finch)

Security checks for vulnerabilities and agentic risk

Overview

The skill uses named services for public-market analysis, requires approval for paid runs, and shows no hidden or unrelated access.

Send only the public-market question and necessary details; omit personal financial information and account secrets. Check free-run availability, approve any paid run explicitly, and review connection permissions before enabling MCP access.

A.I.G

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

SkillSpector

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 forwards the user's input to third-party services 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 task containing that input through Finch for the Powerdrill Finance Data Agent. Although the services are named, the workflow does not warn the user that their question will be transmitted externally or advise against including sensitive financial or personal information.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.