Back to skill

Security audit

HK-101 Living RAG

Security checks for vulnerabilities and agentic risk

Overview

This is a small local-document RAG skill with no bundled code or persistence, but users should understand that using an OpenAI API key may send queries and document snippets to an external provider.

Install only if you are comfortable using an OpenAI-backed workflow for the selected docs. Keep docsPath pointed at a narrow, non-sensitive folder, and avoid private, regulated, credential-containing, or broad home/project directories unless you intend that content to be used in model requests.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly requires an external API key and performs RAG over local text/markdown, but the description does not warn users that snippets or document content may be sent off-host to a third-party model provider. This creates a real risk of unintended disclosure of sensitive local files because users may assume processing is purely local from the phrase 'over local text/markdown.'

Static analysis

No suspicious patterns detected.